Ship Weight Reduction and Efficiency Enghancement Through Combined

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2013 Ship Weight Reduction and Efficiency Enhancement Through Combined Power Cycles Michael J. Coleman

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected] THE FLORIDA STATE UNIVERSITY

COLLEGE OF ENGINEERING

SHIP WEIGHT REDUCTION

AND EFFICIENCY ENHANCEMENT

THROUGH COMBINED POWER CYCLES

MICHAEL J. COLEMAN

A Thesis submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Master of Science

Degree Awarded: Spring Semester, 2013

Michael J. Coleman defended this thesis on April 1, 2013. The members of the supervisory committee were:

Juan C. Ordonez Professor Co-Directing Thesis

Alejandro Rivera Professor Co-Directing Thesis

Farrukh S. Alvi Committee Member

Carl A. Moore, Jr. Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

This thesis is dedicated in gratitude to my parents, Norwood Sr. and Alice Coleman. It is dedicated in loving memory to my grandparents Viola Smith, and Charles Frank Coleman.

This thesis is dedicated to the prosperity of my daughter, Onyame Coleman, with respect to my brother, Norwood Coleman, Jr, and with thanks and gratitude to God.

This thesis is also dedicated to all of my family, and friends, who are too numerous to mention here. This thesis is dedicated to the educators at every level who have influenced my life and career.

Special thanks goes to my boss, Ferenc Bogdan, at the Center for Advanced Power Systems (CAPS), who has been very patient during this process, Steinar Dale, CAPS’ director, all of the facilities staff at CAPS, and all of my co-workers at CAPS.

iii

ACKNOWLEDGEMENTS Partial support for this work from the Office of Naval Research (ONR) and the Naval Engineering Education Center (NEEC) is greatly appreciated. I would also like to acknowledge Alejandro Rivera, Carl Moore, Juan Ordonez and Emanuel Collins, whose support and encouragement has been indispensible, as well as Leon Van Dommelen, and Anter El-Azab, who emphasized the role and importance of mathematics in the to me in the pursuit of scientific concepts.

As always, I wish to acknowledge the love and support provided by Norwood Sr. and Alice Coleman, for their unwavering support at every stage of life, in every way possible.

TABLE OF CONTENTS List of Tables ...... vii

List of Figures ...... viii

Abstract ...... x

1. MOTIVATION AND LITERATURE REVIEW ...... 1

1.1 Combined Cycles for All-Electric Ship Applications ...... 1

1.2 Variations of Hybrid Electric Ship Configurations ...... 4

1.3 Installed All-Electric Ship System ...... 6

2. ANALYSIS OF A COMBINED CYCLE POWER PLANT ...... 8

2.1 Overview of the Combined Cycle Power Plant ...... 8

2.2 The Gas Turbine Prime Mover ...... 10

2.3 The Steam Power Plant ...... 13

2.3.1 The Heat Recovery Steam Generator (HRSG) ...... 13

2.3.2 The Steam Turbine ...... 17

2.3.3 The Condenser and the Pump ...... 21

2.4 Design Strategy for the Combined Cycle Power Plant ...... 23

2.4.1 Combined Cycle Power Plant Configuration Analysis ...... 23

2.4.2 Roadmap to Combined Gas and Steam Turbine Power Plant Configuration ...... 26

2.5. Summary ...... 30

3. WEIGHT ANALYSIS ...... 31

3.1 Weight Considerations for the Combined Gas and Steam Turbine Power Plant ...... 31

3.2 Turbine Weight ...... 33

3.2.1 Gas Turbine Weight ...... 36

3.2.2 Steam Turbine Weight ...... 38

3.2.3 Electrical Generator ...... 41

3.3 Heat Exchanger Weight and Heat Transfer Area ...... 41

3.4 Fuel Volume and Weight ...... 46

3.5 Summary ...... 56

4. ANALYSIS RESULTS ...... 57

4.1 Case I – The Effects of Varying HRSG Exhaust Gas Temperature (T5) ...... 59

4.2 Case II – The Effects of Varying Steam Quality (x8) ...... 68

4.3 Case III – The Effects of Varying HRSG Pinch Point ...... 73

4.4 Case IV – The Effects of Gas Turbine Performance ...... 77

4.5 Analysis Summary ...... 80

5. CONCLUSIONS & FUTURE WORK ...... 82

BIBLIOGRAPHY ...... 85

BIOGRAPHICAL SKETCH ...... 87

LIST OF TABLES

Table 1 – Typical Combined Cycle Model Constants ...... 24 Table 2 – General Electric Gas Turbine Scale Factors Exponents ...... 35 Table 3 – Data for Commercial Gas Turbines ...... 37 Table 4 – Data for Commercial Steam Turbines ...... 40 Table 5 – Weight Distribution of Conventional Turbine-Generators ...... 42 Table 6 – FU Values for the Heat Exchangers in the Power Plant ...... 45 Table 7 – Notional Ship Power Specifications [3] ...... 49 Table 8 – Economical Transit Fuel and Volume Savings ...... 53 Table 9 – Surge to Theater Fuel and Volume Savings ...... 54 Table 10 – Operational Presence Fuel and Volume Savings ...... 54

vii

LIST OF FIGURES

Figure 1 – Major Components of a Combined Cycle Power Plant...... 10 Figure 2 – Energy Flows Crossing the Gas Turbine’s Boundary ...... 11 Figure 3 – Rankine Cycle TS diagram ...... 14 Figure 4 – Heat Recovery Steam Generator ...... 15 Figure 5 – Potential Pinch Point Visualization ...... 16 Figure 6 – Effects of superheated and high quality live steam ...... 19 Figure 7 – Effects of low quality steam turbine exhaust on HRSG design ...... 20 Figure 8 – Condenser and Pump ...... 22 Figure 9 – Logic Flow for Combined Cycle Power Plant Configuration ...... 27 Figure 10 – Power versus Weight for known gas turbine ...... 38 Figure 11 – Power to Change in Enthalpy Ratio versus Weight for known steam turbines ...... 40 Figure 12 – Heat exchanger notation for logarithmic mean calculations ...... 43 Figure 13 – Dry weight of commercially available HRSGs and condensers ...... 46 Figure 14 – Percent fuel weight and volume reduction with increasing power plant efficiency .. 55 Figure 15 – Fuel weight savings with increasing efficiency...... 55 Figure 16 –HRSG exhaust gas temperature is a qualitative measure recovered power ...... 58 Figure 17 – Efficiency versus T5 in the format used to evaluate other parameters ...... 59

Figure 18 – The effects of changing T5 on the steam turbine power output ...... 61

Figure 19 – The effect of reducing T5 on the location of the pinch point ...... 62 Figure 20 – Gross effects on mechanical component, fuel, and net power plant weight ...... 63

Figure 21 – The effects of varying T5 on the net weight ...... 64 Figure 22 – Mechanical Components Breakdown ...... 65 Figure 23 – The effect of HRSG power variation on required heat exchanger surface areas ...... 66 Figure 24 – System weight for a 1/4 range Economical Transit-type mission ...... 67 Figure 25 – Net weight versus for full, 1/2, 1/3, and 1/4 Economical Transit trip durations ...... 68 Figure 26 – Net weight reduction versus efficiency for several Economical Transit-type trips . 69 Figure 27 – Effects of quality on combined cycle efficiency ...... 70 Figure 28 – Effects of quality on combined steam turbine power ...... 70 Figure 29 – Effects of quality on weight savings ...... 71

viii

Figure 30 – Effects of quality on weight impact per day for full range mission ...... 72 Figure 31 – Effects of quality on HRSG and condenser weight ...... 72 Figure 32 – Effects of quality on mechanical component weight ...... 73 Figure 33 – Effects of changing the HRSG pinch point ...... 74 Figure 34 – Effects of changing the HRSG pinch point on steam turbine power production ...... 75 Figure 35 – Effects of pinch point variation on net weight savings ...... 76 Figure 36 – Effects of pinch point variation on net weight savings per day ...... 76 Figure 37 – Efficiency response to various gas turbine prime movers ...... 78 Figure 38 – Steam turbine power output for various gas turbine prime mover configurations .... 78 Figure 39 – Net Weight Savings for various gas turbine prime mover configurations ...... 79 Figure 40 – Net Weight Savings per day for various prime mover configurations ...... 80

ABSTRACT

This work describes a tool for configuring and analyzing the weight of combined cycle power plant, designed for shipboard applications. The effects that varying selected combined cycle parameters have on the weight and the efficiency are presented. The combined cycle configuration is limited to a simple Rankine cycle bottoming plant recovering power from a gas turbine prime mover in order to increase efficiency. Although the Rankine cycle analysis could be used to design a steam turbine cycle whose HRSG absorbs power from the waste heat from any prime mover, the weight analysis provided constricts the use of the tool to gas turbines. Unlike much of the weight analysis performed in contemporary literature, this work includes fuel weight as part of the power plant weight, and the analysis results in net weight savings as compared to simple cycle gas turbines operating alone. The model was developed using heat transfer and thermodynamic analysis, turbine scaling techniques, and data from commercially available hardware to size the major power plant components. The analysis reveals that the point of optimal weight does not always coincide with the point of optimal efficiency.

CHAPTER 1 MOTIVATION AND LITERATURE REVIEW

As the Navy transitions to an all-electric fleet, analytical tools must be developed to facilitate a rational process for the selection of power plant configurations for specific ship designs and modes of operation. Many hybrid designs are being considered for future Navy power plants. These alternatives have been developed in response to the increasing demand of both commercial companies and world Navies to implement all-electric ship (AES) designs. In this work, a tool has been developed that could be used to analyze a combined cycle power plant that uses a gas turbine prime mover and a steam turbine bottoming cycle. The model will be deployed to examine the effects of varying several operating parameters on the power plant’s weight and efficiency. The results of the analysis are presented in contrast to the operation of a simple cycle gas turbine power plant, and used as a tool for optimizing the combined cycle plant for installation on a frigate-sized navy ship; operating in its least power intensive mode of operation.

Section 1.1 provides a brief review of the literature on the topics of combined cycles for marine applications, and also a look at the role of the all-electric ship for future ocean-going vessels. Section 1.2 is an extension of the literature review. A more detailed discussion of options for hybrid-electric marine power plants is presented. The chapter will conclude by considering the effects of a combined cycle power plant installation on a commercial pleasure cruise ship in section 1.3. The findings of that case study will facilitate a brief discussion about the benefits associated with such implementations in the commercial market, and the presumed benefits for such installations on naval ships.

1.1 Combined Cycles for All-Electric Ship Applications

The papers below were very informative illuminating regarding the current state of the all- electric ship, and its future role for the Navy Fleet. In 2008, McCoy conducted a survey of the expansion of electric ship propulsion options since the early 1980s [1]. He indicated that in response to increased electric sensor and weapons loads that are being planned for installation on future Navy ships. He emphasized the movement away from direct drive architecture that has 1

been the staple for traditional ship propulsion. This model requires a separate set of prime movers that are dedicated to the generation of electrical power for ship service loads. In the integrated, all-electric ship architecture, both propulsion and ship service loads are generated and supplied to a main bus at constant voltage and frequency, with variable frequency drives providing the voltage and frequency required by load. The loads on the all-electric ship are both propulsive and ship service, but they provide great flexibility for Naval architects in the design and layout of future ships that is impossible with the segregated plant design.

In 2011, Lundquist looks towards the future an article published in Naval Forces magazine [2]. Quoting Royal Navy Attaché, Ian Atkins, “Going full electric was/is the same step as going from sail to steam.” Lundquist insists that the US Navy is committed to the all-electric ship is the future, due to a continuous rise in electrical demand onboard. In particular, new weapons systems and technologies, such as rail guns and experimental launchers could replace conventional guns. Moreover, experimental, high-powered radar could also draw from the common electrical bus that the propulsion system accesses for power on the all electric ship.

In that Lundquist’ article, he quoted Captain Doerry, who suggests, both in the article, and in his 2007 paper [3] that new standards and ways of sizing power plants for the all electric ship need to be considered. Doerry’s work was immensely useful in this work for both the determination of the combined cycle configuration, as well as the fuel weight calculations. Doerry also collaborated with Robey, Amy, and Petry in a 1996 article in Naval Engineer’s Journal [4], in which the future of the all-electric ship architecture was discussed.

Holsonback and Kiehne published a paper in 2010 [5] that emphasizes the thermal management challenge that all-electric ships face. They produced a simulation that demonstrates the amount of heat generated onboard a ship while conducting a momentum-reversing maneuver. Elsewhere, Ammonia-water absorption refrigeration plants are held up as a good option for converting waste heat into useful energy on marine vessels [6]. They are projected to save 2-4% of fuel consumption.

In this work, the combined cycle power generating system, which is presented as a candidate for power production on the all-electric ship can also serve as a primary thermal management tool. Implementation of combined cycle power plants onboard ocean-going vessels can help in the thermal management of all-electric ships, while providing other advantages.

Haglind’s 2008 paper, that was issued in three parts [7] [8] [9], points out that the power density of combined cycle power plants is a primary motivating factor in their selection over what has traditionally been the prime mover of choice on ships, two-stroke diesel, operating on heavy fuel oil. In this work, Haglind emphasizes environmental and human health concerns in part 1. He provides the case study discussed later in this chapter, as well as illustrations showing the dramatic space savings associated with the implementation of combined gas/steam turbine cycles in place of diesel engines. In part 3, Haglind describes the dramatic reduction in environmental impact that replacing gas turbines as prime movers onboard, opposed to diesels has. Moreover, the loss in efficiency associated with switching from diesels to gas turbines is largely offset by the addition of the steam turbine cycle.

Haglind emphasizes the declining economical advantage that operating diesels on heavy fuel oil has over operating cleaner burning gas turbines throughout his work. This increasing emphasis on environmental stewardship is underscored by Nord and Bolland’s work, which describes the emissions and efficiency benefits associated with combined gas and steam turbine power plant installations, as compared to gas turbines operating alone [10]. This study was motivated by the desire to decrease operating costs related to CO2 emissions associated with Norwegian regulations for off-shored oil and gas installations.

Another advantage that diesels have historically held over gas turbines is in partial load performance. Haglind suggests that combined cycles can also help to bridge this gap in his 2011 work [8].

In fact, Young, Little and Newell, inform the community that world navies are not leading, but rather trailing the commercial fleet in implementing combined cycles ship-board as power plants in their 2001 work [11]. Emmanuel-Douglas provides a detailed description of several options

considered or utilized by the cruise line industry to implement combined cycle power plants [12]. This work is the primary focus of the next section’s discussion.

1.2 Variations of Hybrid Electric Ship Configurations

Ship power plants are required to provide propulsion, ship service, and heating power as needed to the vessel in varying ratios, depending on the ship’s size, type and the prevailing mode of operation [12]. In response to the constantly increasing number of electrical loads onboard contemporary ships, and the desire to increase the ease of power distribution onboard military- type vessels, the contemporary all-electric ship (AES) concepts were presented at ASNE day 1994 [4]. In the AES design all of the power generated onboard is converted into electricity, and made available for either propulsion or ship service use. The transition to electric propulsion essentially serves an alternative power transmission option to a mechanical gear train for converting the high speed, low torque prime mover shaft output to the low speed, high torque shaft response that is required to turn propellers and move large ships [1].

Despite their inherent efficiency disadvantages, the gas turbine has been selected over the diesel engine as the prime mover of the future, primarily because of its high power density [13] [11]. The two primary disadvantages of diesel prime movers are that they consume large amounts of space that could otherwise be used to increase payload capacity, and they are significantly heavier than turbine-powered alternatives. With the selection of gas turbines to power the future all electric (AES) fleet, the massive amounts of waste heat produced by the prime mover must be managed effectively. In order to remove this heat using chillers, massive amounts of cooling infrastructure would be required [5]. This solution would add tonnes of additional equipment, offsetting the gas turbine’s power density benefit over diesel prime movers, while providing no benefit to its efficiency deficit as compared to the diesel.

A more effective option for managing the increased heat load of gas turbine prime movers while enhancing their efficiency is to implement combined cycle power plant technology, which features one or several gas turbine prime movers operating in coordination with a steam bottoming cycle. This alternative is still more power dense than diesel engines, and is capable of producing comparable efficiency [13]. Terrestrial combined cycle power plants that operate in 4

the 100s of megawatts level and are capable of achieving efficiencies of approximately 60%. Ship-board applications are less efficient, because weight restraints necessitate the use of less powerful prime movers, which results in less heat rejection, and subsequently less effective energy recovery [8].

In the past combined cycles power plants have rarely been used for propulsion of ships, but in anticipation of expected legislative action, due to increasing environmental awareness, it is expected that the price of heavy fuel oil (which is used to power diesels at low cost currently) will increase sharply in the coming decades. This projected sea change in world affairs further enhances the prospects of combined cycle systems as viable options for AES designs [13]. Furthermore, emissions reductions associated with switching from diesel cycle prime movers to gas turbine prime movers ranges from 67% for nitrous oxide emissions to 98% for carbon monoxide and hydrocarbon emissions. These emission reductions are partly due to the use of the higher quality fuels burned by gas turbines, and partly due to the different combustion processes [9]. Another advantage that gas and steam turbines hold over diesel engines is an artifact of their higher operating frequencies. Noise and vibrations from turbines are more easily damped. It has also been demonstrated that start-up times for diesel engines and combined cycle power plants are comparable [8].

There are several ship-board designs for combined cycle power plants that are either currently in use, or under consideration for future use in Navy ship installations. The options range from conventional direct mechanical propulsion designs, to hybrid mechanical and electric propulsion options, to fully integrated AES designs [12].

Combined cycle designs that employ conventional mechanically coupled gear trains from turbines to propeller shafts include the combined power and heat generation (COGEN), the combined gas turbine and steam turbine (COGAS), the combined gas and gas turbine or alternatively combined gas or gas turbine (COGAG, or COGOG), and the combined diesel or gas turbine or alternatively combined diesel and gas turbine (CODOG, or CODAG) configurations. Designs that operate on a hybrid mechanical and electric propulsion platform include the combined gas turbine and electric (COGAL), the combined gas electric and gas turbine

(COGLAG), and the diesel engine with waste heat recovery (DE/WHR) configurations. Fully integrated all electric propulsion ship designs can be achieved using the combined gas turbine electric and steam turbine (COGES) or the combined diesel electric and gas turbine (CODLAG) configurations [12]. In addition to the afore-mentioned possibilities, a combined cycle option consisting of a turbo-diesel whose exhaust can be controlled to drive a gas/steam turbine combined cycle and/or provide heat to an HRSG for steam power operation has been proposed [14].

In this work, a COGES-like power plant design has been selected for analysis. While it is clear that AES designs incur additional losses by converting mechanical energy into electrical energy, and then back into mechanical energy, the dramatic increase in power management flexibility has been determined to outstrip the modest efficiency gains associated a direct mechanical coupling from prime movers power to the propellers. For example, an electric propulsion system integrated with the ship service distribution system offers naval architects considerable flexibility and often the choice of a more affordable ship. Electric drive provides flexibility in planning the placement of ship components in the hull. Decoupling prime movers from drive shafts permits location of prime movers to be optimized for maximizing payload carrying capacity [12] [4]. Advantages of implementing the IPS can also be realized in ship manufacturing [4].

1.3 Installed All-Electric Ship System

The world’s militaries trail industry in the implementation of combined cycle power generation systems. The use of a common power system for both propulsion and ship services is now a standard commercial practice for the cruise market and specialized shipping. Efficient operation of combined cycle power plants is being achieved by minimizing the number of prime movers required to meet the mission’s load requirements, and operating the turbines at or near their optimum efficiency [11].

In the year 2000, Celebrity Cruise Lines introduced the world’s first pleasure cruise ship powered by a combined-cycle power plant: the GTS Millenium. The use of a COGES plant to power Celebrity’s ground-breaking Millennium ship design freed up space for 50 additional passenger cabins. The installation also reduces the amount of ancillary machinery required for 6

operation. Similar results have been demonstrated in studies of 2500 other passenger cruise ships [8].

If this type of space savings were applied to a military vessel, ship designers could add munitions, sensory, or other mission-critical equipment to the ship, without increasing the weight of the overall vessel. Traditional Navy power systems require at least four prime movers (two for propulsion and two for ship service load), to comply with redundancy standards. Under many operating conditions, prime movers are idled for lack of demand. However, by feeding the power generated by online prime movers onto a common bus, from which the propulsion and ship service power needs can be pulled on demand, the number of prime movers can be reduced significantly. For example, it has been suggested that a conventional destroyed, that is typically deployed with seven prime movers (four propulsive, and three ship service) could operate with as few as three prime movers if combined cycle power plant designs were implemented [4].

CHAPTER 2 ANALYSIS OF A COMBINED CYCLE POWER PLANT

As the attention paid to the environmental impact of burning fossil fuels, and fuel costs continue to increase, the requirement to maximize the use of the energy available from hydrocarbons has become paramount in the design of combustion-type power plants [12] [7] [8] [9] [13] [14]. In response to these trends, and the Navy’s call for more dynamic access to the power generated on its ships, the options for its future power plants have been expanded to include combined cycle, all-electric ship (AES) architectures [14]. The combined cycle power plants using gas turbine prime movers and complimentary steam cycle to augment power production, serves as the model for the tool developed in this work. As with the development of any power plant designed for transportation, efficiency and weight are critical design parameters. The efficiency of combined cycle architectures can be calculated from information provided in this chapter. That information will be used in the next chapter to determine the weight of the power plants designed for a frigate-sized Navy ship.

Section 2.1 presents an overview of the combined cycle. In sections 2.2 and 2.3, a discussion of modeling techniques for plant components is conducted. Section 2.4 includes a detailed discussion of the solution methodology developed for configuring combined cycle power plants, and section 2.5 is used to summarize the developments of this chapter.

2.1 Overview of the Combined Cycle Power Plant

Combined cycle power generating systems, merge two power plant designs that interface at a common heat exchanger, known as the heat recovery steam generator (HRSG). In the HRSG, heat from a prime mover that would otherwise be lost to the surroundings is intercepted and used as the power source for the secondary power plant. The cycle that consumes fuel and adds heat to the HRSG is referred to as the topping cycle. The cycle that scavenges heat energy from the HRSG to power its processes is called the bottoming cycle [15]. The topping cycle is typically a chemical engine, but it could be any power generating system that produces sufficient energy in the form of waste heat to power another cycle. For example, the heat rejected from a large bank

of fuel cells or solar panels could be used as the power source for an HRSG. Typical bottoming cycle candidates are the steam power plant and the Stirling engine.

Regardless of the choice of topping and bottoming cycles selected, the intent of a combined cycle power plant is to maximize energy extraction from the topping cycle’s energy source. The work generated by the bottoming cycle is complimentary to the power generated by the topping cycle. Therefore, the combined cycle efficiency is always greater than the efficiency of the topping cycle operating alone (assuming that there is no supplementary heating). The efficiency of the combined cycle power system and its relationship to the efficiency of the topping cycle working alone is defined in Eq. (1).

  (1) (2)

Where, is the combined cycle efficiency. The subscripts “top” and “bot” indicate the work power outputs , the heat power inputs , and the efficiencies  of the topping and bottoming cycles respectively. Since the ratio is always positive, Eq. (1) illustrates that the addition of the bottoming cycle always increases the overall efficiency of the power plant as compared to the topping cycle working alone as a simple cycle.

In this work, gas and steam turbine cycle pairs will be evaluated as the combined cycle power plant of choice. Figure 1 shows a schematic representation of the major components of the power plant. The configuration of the gas turbine is almost universal in combined cycles; however, the steam power plant can be designed in a variety of different configurations. Common modifications include the use of reheating, steam extractions, and water pre-heaters, among others. In the case of Figure 1 the configuration corresponds to the simplest steam turbine configuration, a standard Rankine cycle. In the sections that follow, a basic description of gas turbine and steam turbine power plants is presented. Included with the steam power plant

description, special attention is paid to the HRSG, which serves as the interface between the topping and bottoming cycles.

Figure 1 – Major components of a combined cycle power plant

2.2 The Gas Turbine Prime Mover

The gas turbine is one of the most desirable prime movers for use in electric power generation, large-scale commercial and military marine applications. Gas turbine engines, take in fuel from an onboard tank and air from the atmosphere. The fuel is burned in a combustion chamber, and

the products of reaction are exhausted to the atmosphere. The major components of the gas turbine can be identified in the “topping cycle” portion of Figure 1.

In this treatment, it is not necessary to analyze the gas turbine internal operating parameters. Gas turbine performance will be based on existing commercial designs that can be scaled using techniques developed by the General Electric Corporation [16]. However, in this chapter, gas turbine analysis focuses on the energy transfer from the prime mover’s exhaust gases through the HRSG, and into the water of the Rankine cycle. Figure 2 illustrates typical energy transfers required for gas turbine operation. It is assumed that the machine operates in an environment that conforms to the generic International Standards Organization environment for gas turbines, which assumes atmospheric temperature and pressure of 15 °C and 1.013 bar (1 atm), respectively. These environmental conditions are commonly referred to as ISO conditions [17].

Figure 2 – Energy flows crossing the gas turbine’s boundary

The energy balance for gas turbines, can be written as,

(3)

The gas turbine generating capacity can be expressed as a fraction , of the power

produced by the combined cycle . (4)

A good deal of information about∗ the gas turbine operating parameters can be extracted from manufacturer’s data sheets. A typical data sheet for a gas turbine will include exhaust gas temperature leaving the turbine , the electric generation efficiency , the mass flow rate

of gases leaving the gas turbine , and the output power . The heat input to the gas turbine can be related to the turbine’s power through the general definition of efficiency.

 (5) The heat losses shown in Figure 2 correspond to losses through the turbine walls to the environment. Those losses are unable to be recovered by the bottoming cycle. Fortunately, they represent only a small fraction of the total fuel energy . A typical value for is 4% of the available fuel energy. In this study it is assumed that such a value is valid for all generator sets, ε regardless of power level. (6)

The information above is all that is necessary to perform the analysis of the gas turbine section of the combined cycle power plant in this work. Combining Eqs. (3), (5), and (6), the mass flow

rate of the exhaust gases can be obtained as a function of fixed gas turbine parameters and its variable power.

(7) 1ε 2.3 The Steam Power Plant

Steam power plants can be designed to take in atmospheric air and fuel to fire a boiler; however any heat source can be used to boil the working fluid. In a combined cycle power plant the steam turbine power plant absorbs heat from the gas turbine exhaust in the heat recovery steam generator (HRSG). As shown in the “bottoming” portion of Figure 1, the high temperature products of reaction exhausted from the gas turbine are directed into the HRSG to boil the water that drives the steam turbine cycle,

In this work, well-known Rankine Cycle concepts [15] are used to quantify critical features of the bottoming, steam power generation cycle. Water is compressed in the pump (the process from state 9-6), heated through a phase change in the HRSG (the process from state 6-7), expanded and cooled in the turbine (the process from state 7-8), and finally cooled through another phase change back to its original state (the process from state 8-9) in the condenser before re-entering the pump. For clarity, a Rankine Cycle T-S diagram is presented in Figure 3. Models for the components introduced above are described in the subsections that follow, beginning with the HRSG.

2.3.1 The Heat Recovery Steam Generator (HRSG)

Information determined from the gas turbine analysis, in section 2.2, is used to facilitate the analysis of the heat recovery steam generator (HRSG). The temperature of the exhaust gases leaving the gas turbine (entering the HRSG), the mass flow rate of the exhaust gases, and the temperature of the exhaust gases exiting the HRSG all affect the behavior of the bottoming cycle. In this work, once-through heat recovery steam generator technology has been selected for analysis. This design simplifies HRSG design and eliminates the need for steam drums. Further, 13

with proper materials selection, the bypass stack can be omitted. These design selections are important factors in optimizing the weight and volumetric advantages of non-terrestrial combined cycle installations [10].

Figure 3 – Rankine cycle TS diagram

A representative graphic of the HRSG process is provided in Figure 4. In this work, the HRSG is modeled as a counter-flow heat exchanger comprised of three sections: the economizer, the evaporator, and the superheater. Real economizers are designed so that the feed water at the outlet is slightly sub-cooled at full load, but during off-design operations, steam can be generated in the economizer [13]. However, in this work, the economizer receives compressed liquid water from the pump, which is then heated to a saturated liquid (states 6-6a) by extracting heat from the hot gases (states 4b-5). In the evaporator, the saturated liquid leaving the evaporator is heated to a saturated vapor (states 6a-6b) by extracting heat from the hot gases (states 4a-4b). Finally, the saturated vapor leaving the evaporator is superheated in the superheater (states 6b-7) by extracting heat from the hot gases (states 4-4a).

The sequence described above is typical of HRSG models. However, alternative sequences place water in the evaporator in contact with the hottest gases. This is a common practice in fuel-fired boilers and HRSGs with a supplementary fuel source. Conducting an energy balance of the HRSG facilitates computation of the heat input into the Rankine cycle. Equations (8), (9), (10), and (11) quantify the heat flow from the exhaust gases into the HRSG in terms of gas turbine exhaust properties and the Rankine cycle water properties.

Figure 4 – Heat recovery steam generator

(8)

The energy balance described in Eq. (8) can be subdivided into three discreet energy balances for each individual section of the HRSG.

(9)

(10)

(11)

The total heat transferred into the HRSG, provided in Eq. (8), can also be expressed as the sum of the heat transfer rates into the individual components of the HRSG (the superheater, the evaporator, and the economizer).

(12)

Pinch analysis, in which the closest stream-to-st ream temperature difference (the pinch point) is used as the design goal, is common practice with HRSGs. An inverse relationship exists between the pinch point and heat exchanger area. For various heat exchanger types there are several rules of thumb that dictate recommended values. Typical values for pinch point in HRSGs are

between 10 and 30 °C. Excessively small pinch point values, result in heat ∆ exchanger whose surface area is prohibitively large. Conversely, excessively large pinch point values result in poor heat transfer effectiveness. Figure 5 provides a visual representation of the two regions inside the HRSG that could serve as the pinch point in this analysis. The smaller of the two temperature differences between the exhaust gas and the steam is designated the pinch point for the design.

Figure 5 – Potential pinch point visualization

As shown graphically in Figure 5, the temperature of the superheated steam entering the steam turbine must be less than or equal to the temperature of the gas exiting the gas turbine , less the pinch point design value.

(13)

Simultaneously, the temperature of the∆ saturated water entering the evaporator must be less than or equal to the temperature of the exhaust gas leaving the evaporator and entering the economizer less the pinch point design value.

(14)

It is important to emphasize that the value∆ of the temperature in the evaporator is

dependent upon the operating pressure of the HRSG , according to thermodynamic and relationships of pure, saturated substances. The value of affects the values of , , , , and because of Eqs. (13), (14) and basic thermodynamic∆ relationships. This fact is critical to the steam turbine analysis of the next section.

2.3.2 The Steam Turbine

By definition a thermodynamic state requires two independent properties to be identified. In the bottoming cycle, state 8 can be defined by the operating pressure in the HRSG and the temperature of the steam entering the turbine . Similarly, state 9 can be defined by the quality of steam leaving the turbine , and the operating pressure of the condenser .

The selection of states 7 and 8 are both related to equipment design and protection issues. The design intent is to maximize the pressure and temperature of state 7 so that the

most possible energy is available in the steam for conversion into electricity and at the turbine- generator. Increasing and has the effect of increasing the efficiency of the Rankine

cycle, a highly desirable effect. However, materials limitations cap the maximum allowable conditions for the combination of and . This limit sets the boundary of state 7.

State 8 requires a design balance between minimizing the amount of liquid in the saturated steam exhaust mixture and maximizing the efficiency of the Rankine cycle. In contrast to the effect of changing the HRSG operating pressure, decreasing the condenser’s operating pressure/temperature, elevates the efficiency of the Rankine cycle. Unfortunately, since the isentropic efficiency of the steam turbine cannot be substantially altered, decreasing the condenser operating pressure/temperature has the unwanted effect of increasing the amount of liquid water present in the latter stages of the steam turbine. Too much liquid in the turbine will cause damage to this expensive and critical piece of equipment. As indicated by the xmin in Figure 3, a quality below this point must be avoided or the steam turbine could be damaged during operation.

The pressure of the condenser is a design parameter generally set by steam turbine manufacturers. Its value is a compromise between a very high vacuum at state 8, which would have a positive impact on efficiency, and the economics of producing vacuum conditions. A typical value for is 0.1 bar(a). Once the condenser pressure is fixed (by the manufacturer),

the quality of the exhaust steam becomes a critical design parameter.

State 7 is also critical in the design of the Rankine cycle. Selection of this state is complicated by the fact that it is affected by both the HRSG’s operating pressure and the peak

superheated steam temperature . The effects that increasing and decreasing those parameters has on the quality of steam exiting the turbine will be examined for both fixed pressure, and fixed superheated steam temperature HRSG designs below.

In fixed pressure HRSG designs, an excessively high steam quality, or superheated steam design points could result in HRSG peak superheated steam temperatures in excess of the safe operating envelope for gas turbine materials. This condition is illustrated by lines B, C, and D of Figure 6. This is typically not a problem for HRSGs powered exclusively with gas turbines exhaust, because exhaust gas temperatures are not sufficiently high to stress commonly used steam

18 turbine materials. However, a superheated or excessively high quality steam design point (line B) could result in a value for that violates the inequality in Eq. (13). On the contrary, if the designed steam quality is too low, the efficiency of the Rankine cycle will suffer. Line A of Figure 6 illustrates this condition.

Figure 6 – Effects of superheated and high quality live steam

However, if the HRSG exhaust temperature is the fixed design parameter, increasing the quality at state 9 (i.e. allowing superheated steam) to exhaust from the steam turbine, will result in a loss of efficiency in the Rankine cycle, as a result of the required pressure drop to complete the cycle. This effect is illustrated in line B of Figure 7. On the contrary, if the steam quality is too low (line A in Figure 7), the result could be a value for that violates the inequality in Eq.

(14).

The specific work produced by the steam turbine generator can be determined from enthalpies , given the isentropic efficiency of the steam turbine generator  . As with gas turbine and generators, discussed in section 2.2.1, a steam turbine’s efficiency can be obtained from information provided by the manufacturer. The relationship between an isentropic steam turbine (illustrated in Figure 3) that operates between enthalpies is presented in

Eq. (15); represents the exhaust enthalpy of an isentropic steam turbine and that receives steam with enthalpy .  (15)

Figure 7 – Effects of low quality steam turbine exhaust on HRSG design

The selection of condenser pressure has an effect on the overall performance of the steam cycle. The lower the condenser pressure, the higher the overall efficiency. However, in practice, the minimum condenser pressure is limited by the temperature of the cooling water available [13].

The mass flow rate of the water in the Rankine cycle is related to the power produced by the steam turbine generator and the enthalpies at states 7 and 8 by:

W (16)

Recall that the power produced by the combined cycle power plant is split between the gas

turbine generator power , described in Eq. (4), and the steam turbine power , described in Eq. (16).

(17)

With the information provided above, the efficiency of the Rankine cycle  can be described

using the heat recovered from the HRSG, as calculated using either Eq. (8) or (12), and the power output of the steam generator, which can be calculated using Eq. (16).

(18) Further, the efficiency of the power plant can be calculated using Eq. (19), which is a

modified version of Eq. (1).

  (19) 2.3.3 The Condenser and the Pump

The condenser’s design and construction is critical to efficient operation of the Rankine Cycle. The condenser is a vessel in which exhaust steam is condensed by cooling water from a presumably infinite, constant temperature source (the ocean for Navy ships). The condenser modeled here is a surface type condenser. 21

A surface condenser is a closed vessel filled with many small-diameter tubes. Condensing water flows through the tubes, while steam from the turbine flows on the outside of the tube bank. Turbine exhaust steam enters the top of the condenser, and flows down, around, and between the banks of tubes. Liquid condensate drips down to the hot well, where the water is recovered and pumped back into the HRSG. During the condensation process, vacuum conditions are generated inside the condenser. It should be noted that the saturation temperature of the condenser is clearly less than 100 ˚C (recall Figure 3). This fact requires that the operating pressure of the condenser is less than atmospheric, thus condensers operate in a vacuum condition [18].

In the condenser, the exhaust steam is cooled in the process from state 8 to state 9. Like states 7 and 8 in the turbine, the fluid properties at state 6 are driven by an equipment protection consideration. Pumps operate most effectively when handling pure liquids. Subsequently, condensers are designed to release only liquid condensate at state 6 from the hotwell, as illustrated in Figure 8.

Figure 8 – Condenser and pump

Cooling water of sufficient quantity, with a pre-defined temperature drop, flows through the tubes to condense the exhaust steam. Non-condensable gases and air are removed by ejectors. The removal of air (and the oxygen it contains) reduces the possibility of corrosion in the piping and the boiler [18]. Because the minerals found in the cooling water of many condensers (particularly seawater applications) can be corrosive, condenser interfaces and components should be monitored for corrosion.

The mass flow rate of the cooling water running through the condenser can be calculated from the relationship in Eq. (20).

(20)

Changing focus to the feed water pump, it is assumed that the pump operates isentropically. Therefore, the entropy of state 7 is equal to the entropy of the liquid water at the condenser outlet. (21)

This assumption induces very lit tle error into the analysis, as the breakout of state 6 and 7 in Figure 3 illustrates.

2.4 Design Strategy for the Combined Cycle Power Plant

This section is intended to introduce a method, using the concepts and equations discussed earlier in this chapter, for analyzing a number of combined cycle power plant configurations. Each configuration generated by this part of the tool is capable of producing the power required to power a frigate-sized ship.

2.4.1 Combined Cycle Power Plant Configuration Analysis

Nine parameters have been identified as critical to the analysis of a combined cycle power plant. The effects of manipulating these parameters manifest themselves in the efficiency and weight of

the system. In order to perform the analysis of the power plant problem, most of these variables must be held constant while manipulating a selected few. Typical ranges of the parameters selected for analysis are presented in Table 1.

Table 1 – Typical Combined Cycle Model Constants

COMBINED CYCLE DESIGN PARAMETER VALUE

Gas Turbine Efficiency  25-35%

Gas Turbine Exhaust Temperature 450-550 °C 2-5% Miscellaneous Gas Turbine Losses 0.09-0.15 bar(a) Steam Cycle Condenser Pressure Steam Turbine Exhaust Quality 90-100% 75-85% Isentropic Efficiency of the Steam Turbine  HRSG Exhaust Gas Temperature 100-500 °C 1-40 °C HRSG Pinch Point Combined Cycle Power∆ 15-30 MW The gas turbine efficiency  and the temperature of the exhaust gases that provide power to

the HRSG are provided in manufacturer data sheets. The heat losses through the turbine walls to the environment can also be determined using the information supplied by the gas turbine manufacturer, but it is not provided explicitly. Steam turbine manufacturers also typically ε provide the operating pressure of the condenser and enough information to determine the isentropic efficiency of the steam generator . However, steam turbine manufacturers provide only the minimum allowable quality of the steam turbine exhaust . The actual steam quality is left to discretion of the power plant system designer. The HRSG’s exhaust gas temperature and pinch point in Table 1 are also critical to the power plant design. and are design parameters ∆ that must be worked out by design engineers based on the information ∆ provided by the gas and steam turbine manufacturers. Finally, the maximum

power output of the combined cycle power plant must be provided by the customer, to

satisfy the power requirements during operation. W

Recall from sections 2.3.1 and 2.3.2 that bottoming cycle analysis involves complex relationships between HRSG parameters and the steam turbine exhaust quality; these design variables directly affect the power output of the steam turbine. Unfortunately, a combined cycle cannot be evaluated explicitly using the design parameters selected in Table 1 and the equations discussed in this chapter. Instead, the power plant is evaluated using a model that designates three target variables, selected from Table 1, and three control variables, selected from equations above.

Analysis begins by assigning values to all of the parameters in Table 1, and guessing values for the three control variables. The three target parameters are not included in the iterative model analysis. Instead, interim values for the targets parameters are solved for during each iteration of the model by applying the three control variables and the six remaining design parameters from Table 1 to the equations of this chapter. Following each iteration, the control variables are adjusted, based on the precision to which interim target values match the originally assigned target parameters, and reapplied to the equations along with the six fixed parameters.

The steam turbine exhaust quality , the HRSG’s pinch point , and the maximum

power output of the combined cycle power plant were selected ∆ as the target variables for the model. The control variables selected for analysis W were the power output of the steam turbine , the saturation temperature in the HRSG , and the temperature of

superheated W steam leaving the HRSG . The resulting combined cycle power plant configurations will be used for weight analysis in the next chapter.

It should be recognized that the maximum power required of a ship in certain situations

may exceed the maximum power output of the ship’s combined cycle power plant W. The details of the power requirements for a ship that is designed for a variety of missions W are presented by Doerry [3] for a frigate-sized navy ship. He defines the maximum power output of

an integrated power system (IPS) as the sum of the electrical loads required for ship service and ship propulsion .

W W (22)

The implications of implementing a fixed power combined cycle power plant into a system that requires higher maximum power output than the combined cycle can produce are discussed in the next chapter. The remainder of this section will focus on the implementation of the model developed for analyzing combined cycle power plant configurations, using the design parameters of Table 1 and the control variables to determine (among other things) the efficiency of the power plant.

2.4.2 Roadmap to Combined Gas and Steam Turbine Power Plant Configuration

The analysis conducted in this work requires thee control variables be guessed, and subsequently refined and verified in an iterative process to arrive at meaningful solutions to a variety of combined cycle power plant configurations. The control variables selected for use in the model are: the power output of the steam turbine , the saturation temperature in the

HRSG , and the superheated steam W temperature in the HRSG . In order to refine the guess values, three target parameters from Table 1 must be identified and solved for in the model, using the control variables and the remaining parameters of Table 1 to solve for interim target values. The target parameters are: the combined cycle power output , the

steam turbine exhaust quality , and the pinch point . The process Wallows the calculated values for the target parameters to change with the ∆ guessed values, so that the guesses can be refined, until the target parameters fall within a reasonable degree of accuracy to the originally assigned values. Figure 9 is provided as a visual representation of the process by which the analysis is completed. Using information from steam and other thermodynamic and heat transfer tables, in concert with the equations discussed earlier in this chapter, the process deployed for determining the characteristics of an engineered system are provided below.

Figure 9 – Logic flow for combined cycle power plant configuration

1. Assign values to target, fixed and control variables 1.1. Assign values to the constants in Table 1 1.2. Identify the combined cycle power output , the steam turbine exhaust quality ,

and the pinch point as target variables W

1.3. Guess values for ∆the control variables, power output of the steam turbine , saturation temperature in the HRSG , and superheated steam temperature W in the HRSG

2. Determine the values for parameters that can be evaluated using only the information provided in step 1 and thermodynamic/heat transfer tables 2.1. The saturation pressure , the enthalpy of the saturated liquid water , and the

enthalpy of the saturated vapor water in the HRSG , using the value guessed for the HRSG saturation temperature

2.2. The entropy of the saturated liquid leaving the condenser , using

2.3. The average constant pressure specific heat for the gas turbine exhaust , given and

3. Perform intermediate property calculations

3.1. Determine the enthalpy and the entropy of the superheated steam leaving

HRSG, using steam tables, and the value guessed for

3.2. Determine the enthalpy of the compressed liquid leaving the pump , using steam tables, and

3.3. Determine the enthalpy s s and the quality of isentropic exhaust steam exiting turbine, using steam tables, h and x

3.4. Calculate the enthalpy of the P actual exhaust s s steam exiting the turbine using Eq. (15)

3.5. Calculate the mass flow rate of the Rankine cycle , using Eq. (16) h

3.6. Calculate the heat absorbed by the water in the economizer m , the evaporator , and the superheater using the Rankine Q cycle portion of Eqs. (11), (10), Qand (9), respectively Q

3.7. Calculate the total amount of heat recovered from the gas turbine exhaust in the HRSG , using the Rankine cycle portion of Eq. (8) or Eq. (12)

3.8. Calculate Q the mass flow rate of the gas turbine exhaust , using the gas turbine portion of Eq. (8) m 3.9. Calculate the power of the gas turbine , using Eq. (7)

3.10. Calculate the temperature of the gas turb W ine exhaust as it enters the economizer , using the gas turbine portion of Eq. (11) T

4. Calculate the interim values for the target parameters based on the values of the fixed variables, control variables, and the evaluation of parameters in steps 2 and 3 4.1. Quality of the exhaust steam exiting turbine , can be evaluated using steam

tables, and

4.2. Combined cycle power output can be calculated using Eq. (17)

4.3. Pinch point is the lesser of and . Eqs. (13) and (14) must be satisfied ∆

5. Compare the interim values calculated in step 4 to the originally assigned values for the target parameters 5.1. If the difference between the original and calculated values is greater than the specified tolerance, adjust the guess values and continue at step 2 5.2. If the original and calculated values match, the analysis of the configuration is complete

The assignment of guess values to the control variables in step 1 allows for analysis of a combined cycle power plant’s configuration without rigorously defining every critical piece of information prior to beginning the analysis. This approach to system analysis allows for relatively easy manipulation of critical parameters, using readily available computational tools. Analysis of the system requires a good understanding of the operational processes and access to material properties. The steps laid out above provide a clear procedure for determining the power lever requirements of the major components of the combined cycle system.

With the information determined from the model, many aspects of the combined cycle can be calculated directly, including (but not limited to):

 the efficiency of the Rankine Cycle (steam power plant), using Eq. (18)  the overall efficiency of the combined cycle power plant, using Eq. (19)

The implementation of the previously described algorithm was done in MATLAB. Step 5 was achieved using the FSOLVE function, which adjusts the guessed values based on an iterative quadratic successive method to arrive at the solution.

2.5. Summary

In this chapter, the components and modeling techniques for analysis of the combined gas and steam turbine power plants were presented. Special attention was given to the components of the steam power plant. In particular, the heat recovery steam generator (HRSG), and the steam turbine were analyzed in great detail. However, analysis could not have been completed without performing energy balances on the gas turbine prime mover, the condenser, and the feed water pump. The chapter concluded with a detailed description of the solution methodology of the configuration portion of the tool developed for combined cycle analysis. The information produced from this portion of the tool will be useful for analyzing the weights of the power plant components in the next chapter.

It will be demonstrated that adding weight to the power plant in the form of steam cycle components is more than offset by the combined fuel weight savings achieved by increasing the overall efficiency of the power plant, and downsizing gas turbine prime mover.

CHAPTER 3 WEIGHT ANALYSIS

Unlike terrestrial systems, where the weight of the components is a secondary consideration; the component weight of combined cycle power plants selected for transportation (i.e. marine and aerospace) installations is a key design criterion. The weight of machinery and fuel required to operate the power plant reduces the mission-critical payloads for which these weight and space- limited vessels are deployed. The focus of this chapter is quantification of the weight for the major components of the combined cycle power plant. The power plants considered below are based on the analysis performed in the previous chapter.

In section 3.1, the need for such analysis is expanded upon, and the primary components to be considered are discussed. Section 3.2 focuses the analysis on the gas and steam turbines to be deployed on the ship. A scaling technique developed by General Electric Corporation for gas turbines is used and extended to analysis of steam turbines to estimate the weight of these devices. In section 3.3, the focus shifts to the weight of the heat exchangers required to extract energy from the gas turbine exhaust gases, and to complete the Rankine bottoming cycle. Section 3.4 demonstrates the feasibility of the total system by showing that the weight and space savings achieved by increasing the efficiency of the power plant is well in excess of the weight and space gains required to retrofit a ship with a bottoming cycle.

The methods to be used for analysis of the size and weight were developed to dovetail nicely with the power designations proposed for future all-electric ship designs. The information provided in this chapter together with the configuration analysis discussed in chapter 2, will be used optimize the power plant. These results will seek to find the proper balance between increased efficiency and power plant weight.

3.1 Weight Considerations for the Combined Gas and Steam Turbine Power Plant

In this work, the combined gas and steam turbine power plant is considered as an alternative to simple cycle gas turbines. In the previous chapter, the efficiency advantages of using a combined

31 cycle were discussed. The efficiency gains imply lower fuel consumption and a reduced environmental impact. Although augmenting a simple cycle gas turbine with a bottoming cycle increases the efficiency of power generation, such a change also implies increased cost and complexity. Size and weight are the focus of this chapter. Cost analysis is beyond the scope of this work. Determination of weight for the major components of the configuration using commercially available equipment is the method selected for analysis here.

For nautical and aerospace applications, size and weight are critically important parameters. All of the weight associated with the power generation system reduces the cargo, munitions, and/or payload carrying capacity of the vessel. The issues of size and weight are addressed in this chapter, because they are important to the functional design of navy ships. The energy required to move the ship increases with weight. Moreover, a ship’s density must be limited to maintain buoyancy. Unfortunately, there are no known standardized methodologies available for performing the analysis of power plant size and weight for non-terrestrial installations. In this chapter, a methodology will be proposed. The fundamental quantity for analysis will be the weight. Decomposition of the total weight associated with the power plant can be achieved by use of the following equation:

(23)

It should be noted that is being used to describe the power while describes the weight.

The context should make it obvious which one is being referred to.

Similar analytical approaches will be used to calculate the weights of the gas turbine and the steam turbine , since both devices are the rotating machines that act as prime movers to produce the power required to operate the ship. Device pairs determined from the previous chapter’s analysis will be used to estimate the weight of the required work producing equipment in a combined cycle power plant. Improvements in the power density of the individual devices are beyond the scope of this work.

Likewise, similar analytical approaches will be used to calculate the weight of the heat recovery steam generator and the condenser , since they are both heat exchangers. The

primary factor for consideration with heat exchangers is the surface area through which heat power is transferred from the hot to the cold fluid. After having determined the required heat exchange area, correlations to the weight of the required heat exchanger will be developed from data available for existing heat exch5angers.

Fuel weight is also considered in this analysis, because it represents a significant portion of

the total weight of the ship. Weight analysis of a typical frigate places the power plant and the fuel weights at approximately 6% and 14% of the total ship weight respectively [19]. This indicates that the fuel weight is more than twice the weight of the mechanical power system components in a conventional ship. By increasing the efficiency of the power generation system, the fuel weight required for a given mission is decreased. It will be shown that fuel weight savings can easily offset any increases in equipment weight. Moreover, the weight and space savings could be filled with other mission-critical hardware.

It is very important to understand that this work represents a first approximation of the net weight of a combined cycle power plant compared to a simple cycle gas turbine plant. The analysis does not directly include estimations for the weight of auxiliary equipment such as pumps, pipes, fans, the water in the bottoming cycle, controls and monitoring equipment that are essential to the operation of the combined cycle. These components together are designated as in Eq. (23). To deal with these components, a constant percentage between 3% and 7% of the total weight estimated for the equipment above will be included to compensate for such auxiliaries.

(24)

3.2 Turbine Weight

A scaling technique developed by General Electric (GE) [16] for predicting size, weight, and other features of novel gas turbines, based on information from an existing machine will be

employed as the basis for predicting the weight of both the gas and steam turbines in this work. The power required for the scaled machines will be taken from the device pairs suggested by the model presented in the previous chapter.

A summary of the features highlighted by GE for the scaling gas turbines can be seen in Table 2. GE refers to this design practice as “geometric scaling”. The idea is that the physical size of a machine can be modified while maintaining aerodynamic and mechanical similarity in both the compressor and turbine sections of the gas turbine by increasing or decreasing the rotational speed of the machine. In this work, this approach is extended to steam turbine for analysis of its weight. The scale factor is defined as the ratio of the scaled machine’s diameter to the

diameter of the original machine .

(25)

The value for the exponent presented in Table 2 for various turbine parameters, relate the

designated parameter to the analogous parameter in the original machine by the following relationship:

(26) As expressed in Table 2, application of the geometric scaling techniques results in the pressure ratio in the compressor, the resonant frequency, the mechanical stresses of the machine, the air velocity, and the tip speeds of the rotating components of the novel machine remaining constant. The angular velocity of the rotating shaft is inversely proportional to the scale factor because the angular velocity of the machine must be inversely proportional to the machine diameter, in order to maintain constant velocity and tip speeds on the turbine blades. The unity relationship between efficiency and scale factor agrees nicely with the assessment from the previous chapter that the efficiency will be constant (recall Table 1). The weight of the machine varies with a cubed scale factor , because the weight is proportional to the volume, which is a product 1: 34 of the diameter squared, and the length. In order to maintain mechanical similarity, the length must vary in direct proportion to the diameter.

Table 2 – General Electric Gas Turbine Scale Factors Exponents

Turbine Parameters Scale Factor Exponent Pressure Ratio 0 Frequency 0 Stresses 0 Velocities 0 Tip Speed 0 Angular Velocity (rpm) -1 Efficiency 0 Weight 3 Mass Flow Rate 2 Power 2

When considering the power and the mass flow rate, it is important to recognize that steady state operation is assumed, and that the scaling presumes that the air flowing through the machine is an ideal gas. With these assumptions, it can be deduced that since the pressure ratio is 1:1, and aerodynamic similarity requires that the density of the air must have 1:1 proportionality at any given point in the machine. The temperature must also have a 1:1 relationship at similar positions in the model, because of the ideal gas relationship . Since the mass flow rate of air through the gas turbine is equal to the product of density, velocity, and area, and area is proportional to the diameter squared, it is easy to see that the mass flow rate must vary with the squared scale factor . Finally, the power is directly proportional to the mass flow rate, which is corroborated by the relationship between power and mass flow rate defined in Eq. (7). 1:

Equations (27)-(29) define the scaled models power, mass flow rate, and weight as functions of the scale factor and the original machine’s power, mass flow rate, and weight, respectively.

(27) (28) ∗ (29) ∗ ∗

Solving for the scaling factor in Eqs. (27) and (29), and setting the results equal to each other; a relationship for the turbine weight as a function of the power can be expressed a follows:

(30)

The determination of the coefficient is particular to the machine selected. Evaluation of the

gas turbine-specific coefficient andk the steam turbine-specific coefficient is included in the discussion of the next two k sub-sections. k

3.2.1 Gas Turbine Weight

Equation (30) can be used to extrapolate the weight of a scaled gas turbine from an existing model of known weight and power generating capacity. In order to calculate the weight of the gas turbine for any configuration determined from the analysis performed in chapter 2, a gas turbine-specific coefficient for scaling the gas turbine must be calculated.

k Consider a gas turbine with sufficient power to supply 100% of the propulsion and ship service needs of a ship. The weight and the power of the selected machine can be read from manufacturer’s specifications. w The gas turbine-specific W coefficient for such a machine is:

(31)

The data presented in Table 3 shows the power, weight, and factor for a number of commercially available gas turbines. The GE models are all marine-specific k turbines. The data indicates that as the weight increases, the value for decreases. All of the values for fall between 0.15 and 0.35 . k k Table 3 – Data for Commercial Gas Turbines

Gas Turbine Power Weight Manufacturer & (MW) (tonne) Model Number GE – LM 500 4.5 2.8 0.294 054 GE – LM 1600 14.9 10.9 0.189 292 GE – LM 2500 25.1 20.6 0.164 519 Solar – Centaur 50 4.6 3.2 0.324 349 Solar – Taurus 60 5.7 3.2 0.235 146 Solar – Taurus 70 8.0 4.2 0.185 616 Solar – Titan 130 15.0 9.6 0.165 247

Consider the Solar Centaur 50, and Taurus 60 machines. Although the Centaur produces 1.1 MW less power, the weight of the two machines is identical. The reason for this is that regardless of the power of the machine, some weight is required for the mounting structure. This weight will not change significantly with the power output of the machine, but as the power output decreases, the non-power generating structure contributes a significant portion of the machine’s weight. Taking this into consideration a better way to approximate the weight of the gas turbine as a function of the power is:

(32)

Figure 10 shows the data from Table 2, plotted along with graph of Eq. (32), using a value of

0.150 tonne per and a value of 1.5 tonne. The graph shows good agreementk between the data provided and the curve generated across the entire range of gas turbines provided. MW

Figure 10 – Power versus weight for known gas turbine

3.2.2 Steam Turbine Weight

Before calculating the steam turbine coefficient , the validity of extending the similarity assumptions to the steam turbine must be examined. Equation (16) reveals that the power in the k steam turbine is directly proportional to the product of mass flow rate and the enthalpy drop across the turbine. The similarity assumptions for gas turbines assumed that the change in enthalpy will be proportional to the mass flow rate for scaled machines. Steam turbines, however, are typically designed to operate over a wider range of pressures and temperatures, which allows a single steam turbine to operate over a much wider power range than gas turbines. However, this design feature disrupts the similarity assumptions used in the General Electric (GE) model. In order to use the incorporate the GE model for steam turbine scaling, a modification of the k-value must be incorporated to account for the potential lack of similarity in the enthalpy drop across the machine.

Assuming that the relationship between weight and mass flow rate in Eqs. (28) and (29) remains valid, replacing mass flow rate with its equivalent power to change in enthalpy relationship from Eq. (16), and performing a process similar to the one used to create Eq. (30), the following expression is obtained.

(33) where ∆h

(34) w ∆h With the scaling coefficient for the steam turbine specified, the weight of a particular machine can be determined by applying the steam turbine coefficient, and the power generation requirement, to Eq. (33).

The data presented in Table 4, shows the power range, weight, change in enthalpy, and factor for several commercially available steam turbines. The data indicates an average k value of 0.356 . However, as was the case with Eq. (32) and Figure 10, a modified versionk of Eq. (33) is used to generate the line in Figure 11. (35)

The values used for and in Eq. (35) are 0.232 786 and 1.811 tonne, respectively. m k The line was generated to pass through the points represented by the Elliot and the Siemens machines, because they showed nice agreement in values.

39 k

Table 4 – Data for Commercial Steam Turbines

Change in Steam Turbine Power Weight Enthalpy Manufacturer & (MW) (tonne) Model Number ELLIOT MYR (4‐stage) 3.730 4.310 975 0.575 998 ELLIOT MYR (6‐stage) 5.222 4.535 975 0.365 872 ELLIOT MYR (8‐stage) 7.5 7.71 975 0.361 384 ELLIOT MYR 2SQV6 11 11.5 773 0.214 063 Siemens SST 300 20 26 938.9 0.264 458

Figure 11 – Power to change in enthalpy ratio versus weight for known steam turbines

It should also be recognized that large excursions from the original machine characteristics, will result in increasing error in the approximation of the novel turbine weight, based on Eq. (36). It is expected that the steam turbines selected for the combined cycle will generate between 20% and 40% of total power plant capacity. Smaller machines will likely not be attractive, as they will not provide a significant increase in overall efficiency, and larger machines push the limits of what is feasible for cogeneration power production by a steam generator.

3.2.3 Electrical Generator

The discussion above covers the conversion of the chemical energy in the fuel into mechanical power at the shaft of the turbines. For the all-electric ship, that shaft energy must be converted into electrical energy for distribution throughout the ship. The conventional way to convert the shaft power of a turbine into electrical energy is to decrease the speed of the shaft through a mechanical gearbox, then use a generator to produces alternating current electricity with a desired frequency (typically 60 Hz for U.S. vessels). Additionally, the gearbox requires an oil system, and the generators and turbines require control systems for proper operation. All of these things add weight to the system. In fact, as shown in Table 5, the gas turbine weight represents only, approximately one tenth of the total system weight of a gas turbine-generator package. However, there are opportunities to change the weight distribution of the power generating system. One option is to directly couple high speed generators to the shaft of the turbine, and deploy converters to modulate the electrical power to the desired voltage and current. This option would eliminate the gearbox and its oil skid, as well as significantly reduce the size of the generator. However, the details of such analysis fall outside of the scope of this work, which is to optimize the size and weight of machinery required to convert fuel energy into the shaft power available to the selected generation system. Weight optimization of the system used to convert the shaft power into electrical power, and the subsequent distribution is left to future analysis.

3.3 Heat Exchanger Weight and Heat Transfer Area

For the purpose of this analysis, it is assumed that both the heat recovery steam generator (HRSG) and the condenser are considered counter-flow, shell-and-tube heat exchangers. An analytical method has been selected for the prediction of heat exchanger weight. The weight of a

designed heat exchanger will be projected from information available about the weight of commercially available heat exchangers of the same type.

Table 5 – Weight Distribution of Conventional Turbine-Generators

Total System Manufacturer Gear Box Turbine Weight Generator Weight Model Number Weight (ton) (tonne) Weight (tonne) (tonne) Solar Centaur 50 1.4 3.2 11.7 35 Solar Taurus 60 1.4 3.2 13 35 Solar Taurus 70 3.2 4.2 19 55 Solar Titan 130 5.3 9.6 29 72

The fundamental equation required to determine the required surface area of a heat exchanger is: (36)

where is the total heat transfer ∆ rate for either the heating or cooling fluid as it traverses the heat exchanger. is the correction factor for the heat exchanger, which depends on the particular arrangement of the tubes and shells and which could be found in charts. The correction factor is typically 1 for counter flow heat exchangers, and less than one for heat exchangers that are not counter flow designs. is the overall heat transfer coefficient for the heat exchanger under

consideration, and is the surface area of the heat exchanger tube through which heat transfer takes place. Finally, the logarithmic mean temperature difference of the heat exchanger contrasts the temperatures of the hot and cold fluids as they enter or leave the heat exchanger ∆ [20].

For a cross-flow heat exchanger, the logarithmic mean temperature is given by

(37) ∆ 42 where the temperatures with the various subscripts represent temperatures at specific points in the heat exchanger. Subscripts “i” and “o” represent the fluid flowing into or out of the heat exchanger fluids 1 and 2 respectively [20]. See Figure 12 for a graphical explanation of the subscript notation used in the logarithmic mean temperature calculation.

Figure 12 – Heat exchanger notation for logarithmic mean calculations

In chapter 2, thermodynamic analysis was employed to calculate the heat transfer rates in the HRSG and condenser. Now, heat transfer analysis of the heat exchangers will be employed to determine the area of the heat exchangers, leveraging the heat transfer rates calculated in that chapter. While the heat transfer rate from Eq. (20) can be used directly for analysis of the condenser, the HRSG must be split up into its component parts for analysis. Equations (9), (10), and (11) can be used to determine the values of heat transfer for the superheater, evaporator, and economizer sections of the HRSG respectively. The temperatures of the exhaust gases and the cooling water required for calculation of are available from Eqs. (9), (10), (11), and (20),

∆ 43

but temperatures of the water in the steam cycle must be determined from the enthalpy information in those equations, and a second property (i.e. pressure), using steam tables.

A value for the product can be determined from manufacturer’s data for each heat exchanger.

Average values for FUa sampling of condensers, economizers, evaporators, and super heaters are presented in Table 6. With all the previous information, the areas for the bottoming cycle FU heat exchangers can be determined by using Eq. (36), and implementing the definition for the logarithmic mean temperature from Eq. (37). The particular equations for calculating the areas of the specific heat exchangers are as follows:

(38) (39) (40) (41) It is important to recognize that the HRSG area is the sum of the areas calculated for the economizer, the evaporator, and the super heater.

(42)

The HRSG and condenser areas can be converted into their respective weights by correlating the information provided in Figure 13, which display the dry weight, in tons, as a function of the heat exchanger surface area for several commercially available HRSGs and condensers. Equation (41) is representative of the linear regressions of the data in Figure 13. (43)

Where C represents the slope of the line, A is the heat exchanger area, and is the weight of the smallest support structure for a heat exchanger. Equation (43) demonstrates good agreement with the data. It will be used as a first order approximation of a heat exchanger weight required to implement the bottoming cycle in the next chapter.

Table 6 – FU Values for the Heat Exchangers in the Power Plant

fU Value Device Condenser 3,550-3,940 Economizer 30-40 Evaporator 26-65 Super Heater 82-84

As with any structure, the weight is driven by the volume and density of the materials selected for conduction heat transfer between the fluids. For pipes of constant diameter and wall thickness, commonly used in the manufacturing of HRSGs, the volume of steel is proportional to the heat transfer area. This is the reason for the linear relationship between area and weight in Eq. (43). As was the case with the turbines in the previous section, heat exchangers with zero surface area will not have zero weight. If no surface area is necessary to facilitate heat transfer, then no heat exchanger is required for the application. However, if a heat exchanger is required, the surface area required for heat transfer must be supported by some structure of finite mass. The mass of the structure is not part of the heat exchange surface area, but contributes to the

overall weight of the device. This is the reason for the non-zero crossing of the line proposed for heat exchanger weight estimation formula. In Figure 13 shows values for C and for HRSG’s and condensers as indicated by the diamond and circular dots, respectively. The line generated for the HRSGs has values of 0.009 and 3.872 for C and . The condenser’s values are 0.0126 and 9.653, respectively.

Figure 13 – Dry weight of commercially available HRSGs and condensers

3.4 Fuel Volume and Weight

When analyzing turbine weight, in the previous section, it was obvious that a reduction in turbine size would reduce the weight of the power plant. It was also obvious that adding bottoming cycle equipment to the power plant would increase the weight of the power plant. While it is obvious that increasing the efficiency of the power plant will reduce the weight of fuel required to operate the ship, it is not obvious that the reduction in fuel weight, as a result of relatively modest efficiency gains, is on par with the weight of a gas turbine. The focus of this section is to provide some information to help quantify the scale of fuel weight reduction as a result of the efficiency gains associated with adding a bottoming cycle to the power plant.

The weight of fuel is an important component of a ship’s power generating system’s weight. Conventional ships with segregated propulsion and ship service (or electrical) power plants calculate fuel capacity by specifying an endurance range and an endurance speed for the ship’s mission, and subsequently sizing the fuel capacity for prime movers for propulsion and ship service sub-systems separately. The conventional approach is taken, because the power plants generating energy for propulsion and ship service are independent of one another in conventional ships. As a result, in segregated power system ships, the fuel capacity for the mechanical propulsion power is derived from the endurance speed requirement, while the ship’s electric power generating capacity is determined by the worst case operating condition during the mission. The required fuel calculations are additive and independent of the ships aggregate speed and functionality requirements for a specific mission [3].

However, in the Integrated Power Systems (IPS) proposed for contemporary and future ships, both propulsive and ship service loads are powered by a common bank of prime movers. This power generation architecture offers ship designers increased design flexibility in the consideration of the number of prime movers required on a ship with a defined mission. Contrary to the power requirement calculus of conventional ships, the IPS allows ship power system designers to consider the fact that these vessels rarely, if ever, operate at both maximum speed and maximum ship service output power. In the IPS, there are three primary contributing factors to the required quantity of fuel carried onboard. Those criterion are: the distance that the ship is scheduled to travel on its mission (its range), the speed profile that the ship is expected to operate on its mission, and the electrical loads (both propulsive and ship service) that the power generation system must supply. Obviously, the farther the ship is required to travel, more fuel will be consumed. Similarly, but even more critically, the amount of propulsion power required to move a ship is roughly proportional to the cube of the speed [3].

The use of specific operating modes where it is most advantageous while minimizing its use in either operating regime is and approach used in navy propulsion, where one engine is used for low power cruise operations, while a second engine, with higher power, is engaged for high speed (or boost) operations. In such an arrangement, the smaller engine is either a diesel or gas turbine, while the boost engine is invariably a gas turbine, to accommodate rapid start and high

power density requirements [12]. In some applications, where a vessel’s mission profile requires spending considerable periods both at sea and in estuary, two engines (one of lower and one of higher power) may be arranged to operate as independent propulsion units; either ready for deployment, depending on the mission needs at the moment. Estuary operations in some environmentally sensitive areas would favor the use of gas turbines because of the lower emissions [12]. This section will be based on the methodology suggested by Doerry for military vessels. Three modes of IPS military vessels operation that could be used to determine the size of the fuel tanks are [3]:  Surge to Theater is the mode in which the ship travels as quickly as possible to the theater of deployment. The maximum number of refueling stops allowed to transit a given distance (typically 4000-10,000 NM) at the design speed is part of the surge to theater specification. Refueling is assumed to occur when 50% of the fuel capacity is consumed.  Operational Presence is the mode in which the ship has arrived in the theater and is tactically engaged in its mission. A given speed-time profile and mission capability is defined for the engagement, and a maximum of 1/3 of the fuel capacity can be consumed in this mode.  Economical Transit is most common mode of operation for military marine vessels, as it covers all transportation that is neither Surge to theater, nor Operational Presence. To define this mode, all of the fuel capacity is permitted to be consumed, less the tailpipe allowances. Information associated with the three modes of operation is available for review in Table 7. From this information, it can be determined how much fuel is required onboard to accomplish any mission, given the efficiency of the power plant. Recall that the calculations of chapter 2 define the efficiency for any given combined cycle power plant configuration. Recall from Eq. (22) that the maximum power required for the ship during any mission is the sum of the

ship service and propulsion loads. This value is not the same for the three operational modes W discussed above. WIt varies from as little as 22 MW for Economical Transit up to 64 MW for Surge to Theater. The maximum power required during a mission is defined as the sum of the ship service loads and the propulsion loads.

The assertion that ships rarely operate at maximum speed and maximum ship service output power simultaneously is underscored by the operating speed profile a typical DDG 51 (Arleigh Burke class destroyer). Such a speed profile is typical of the notional ship considered in Doerry´s work. The notional ship typically operates below the 20 kt, the design point specified for Economical Transit mode. Less than 5% of the speed profile for the DDG 51 is represented by the ship operating at 30 kt, which is the maximum operating speed of the ship, and corresponds with the Surge to Theater speed of the notional ship in Table 7. Under most operating conditions, significant quantities of power that are built into the ship’s generation capacity for propulsion can be utilized for mission-critical operations when operating at low speed.

Table 7 – Notional Ship Power Specifications [3]

Surge to Operational Economical Theater Presence Transit Speed (kt) 30 10 20 Range (NM) 4200 n/a 4200 Ship Service Loads (MW) 4 30 4 Propulsion Loads (MW) 60 2 18 Mission Duration (hr) 140 168 210 Maximum Fuel Depletion (%) 50 33 100 Maximum Number of Refueling Stops 2 0 0

In an effort to optimize the power density as well as the operational efficiency of the ship, Economical transit mode has been isolated as the mode for which the combined cycle system will be analyzed. Surge to Theater and Operational Presence modes of operation are mission- critical modes of operation. In these operational modes, mission success outstrips the desire for power plant efficiency. Power density, not efficiency, dominates the design criterion for these power plant configurations. Furthermore, since Surge to Theater and Operational Presence are not the primary modes of operation, it is illogical to tie the base power production machinery to the high power requirements of these modes. The power required to accomplish the other two missions, in excess of what will be produced by the power plant configured for Economical

Transit operation can be satisfied by adding gas turbines to the ship. These machines can be quickly brought online in case a Surge to Theater or Operational Presence is required, but hold a small footprint and less weight than additional bottoming cycle equipment during normal operations, where the additional power is not needed. Moreover, since the IPS will absorb power from every power generating unit and distribute it to electrical devices as needed, the efficiency gains from the combined cycle power plant will elevate the efficiency of the overall power plant in all operational modes.

The information above was presented to provide a framework for the equations below that are used to describe a method for determining the volume and the weight of the fuel

required at embarkation. In addition to the information in Table 7, the driving parameters for sizing the ship’s fuel tanks are: the efficiency of the power plant , defined in Eq. (44), the

low heating value of the fuel per unit volume , the density of the fuel  , and the maximum time at sea before refueling , defined in Eq. (44). It is imperative to understand that the time at sea before refueling is different from the mission duration, which is provided explicitly in Table 7. However, all of the information required to determine the maximum time at sea before refueling is available in the table.

(44)

The definition for overall power plant 1efficiency ∗  can be expressed as a function of the combined cycle power , the combined cycle efficiency  , the maximum power , and the efficiency W of the gas turbine generators  . The expression for the overall power plant efficiency is presented below.

   (45) ∗ ∗ 50

Given the overall power plant efficiency and maximum time at sea before refueling, the maximum electrical energy required to complete the mission can be calculated in megawatt hours. Subsequently, the fuel energy required to supply all of the potential electrical loads during each leg of the mission can be calculated, using the efficiency of the power plant. (46)

∗ (47)  The volume of fuel required to produce the fuel energy can be calculated using the low heating value of the fuel per unit volume. The weight of the fuel is simply the product of the volume and the density of the fuel.

(48) (49)  Consider the three operational modes presented in Table 7, powered by a power plant whose fuel has LHV and density values of 130,000 and 0.9 , respectively. These values are typical of the fuel oils used for marine applications. The fuel weight and space savings can be estimated by assuming that 100% of the ship’s power requirements for Economical Transit operation are satisfied by a combined cycle power plant. In the examples below, combined cycle efficiencies of 40% and 50% will be compared to ship simple cycle power plants operating at just 25% efficiency. Power required above the 22 MW level will be provided by adding 25% efficient gas turbines, similar to the topping cycle prime mover.

First, consider the Economical Transit mode. The data in was generated by contrasting three different power plant efficiency  , values in Eq. (45). The simple cycle gas turbine efficiency   was used to provide reference points for contrasting the volume and

weight savings attainable by deploying 40% and 50% efficient combined cycle power plants (combined cycles 1 & 2, respectively). Using Eq. (44) it is clear that the maximum time at sea is 210 hr. When Eq. (45) is used to calculate the efficiency of the power plant in this mode, the efficiency of the power plant is equal to the efficiency of the combined cycle, according to the power plant description above . The maximum energy required to complete mission is calculated using Eq. (46), the fuel en ergy required at embarkation can be determined using Eq. (47). The volume and weight of the fuel can be calculated using Eqs. (48) and (49), respectively. Table 8 shows the energy, weight and volume savings attainable by increasing the efficiency of the power plant while operating in Economical Transit mode. Next, consider the effects of the combined cycle in the other operational modes. Begin by selecting the appropriate values for the mode under consideration from Table 7; the procedure described above for Economical Transit mode can be followed, to calculate the volume and weight savings attainable in Surge to Theater and Operational Presence modes. However, recall that the efficiencies of these modes are not equal to the combined cycle efficiency, as was the case in the calculation above. Using Eq. (45), the efficiency of these power plants is less than that of the combined cycle power plant used, because of the need for additional, less efficient gas turbines to be brought online, to satisfy the maximum power requirements of these modes. Care should also be taken when calculating the time at sea for Surge to Theater mode with Eq. (44), because unlike the other two modes, refueling stops are scheduled into this mode. The results for Surge to Theater and Operational Presence modes are presented in Table 9 and

Table 10, respectively. Recall that 22 MW of power is supplied by both “Combined Cycle 1” and “Combined Cycle 2” at 40% and 50% efficiency, respectively. Table 9 and

Table 10 show how increasing the efficiency of a base mode of operation can significantly benefit the power plant in all modes of operation.

Table 8 – Economical Transit Fuel and Volume Savings

Gas Turbine Only Combined Cycle 1 Combined Cycle 2 Power Plant Efficiency (%) 25 40 50 Time at Sea (hr) 210 210 210 Max Electrical Energy (GWh) 4.620 4.620 4.620 Fuel Energy (GWh) 18.480 11.550 9.240 Fuel Weight (tonne) 1651 1032 825 Fuel Volume (gal) 485,171 303,232 242,586 Fuel Energy Savings (GW) n/a 6.930 9240 Fuel Weight Savings (tonne) n/a 619 825 Fuel Volume Savings (gal) n/a 181,939 242,586

Recall from Figure 10 that a 25 MW gas turbine weighs less than 20 tonne. The worst case scenario for fuel savings presented in the tables 8-10 is 365 tonnes (Surge to Theater: Combined Cycle 1 + Extra GT), which represents the equivalent weight of roughly fourteen 25 MW gas turbines. In contrast, the power produced by adding two 25 MW gas turbines operating at 100% capacity to the 22 MW combined cycle for Economical Transit operation is more than enough to satisfy the most power intensive mode of operation described by Doerry (Surge to Theater). The levels of space and weight described in Table 9 and

Table 10 will allow ship designers to pack more of the mission-critical equipment inside the ship.

Finally, consider three generic power plants using the same fuel described in the example above, with base gas turbine efficiencies of 25%, 30%, and 35%. Figure 14 illustrates the ratio of combined cycle to simple cycle fuel requirement at embarkation, as a function of the overall power plant efficiency. Figure 15 shows the fuel weight savings in tonnes per GWh of fuel energy required for the mission versus overall plant efficiency.

Table 9 – Surge to Theater Fuel and Volume Savings

Gas Turbine CC1 + CC2 +

Only 25% GT power 25% GT power Power Plant Efficiency (%) 25 30.2 33.6 Time at Sea (hr) 93 93 93 Max Electrical Energy (GWh) 5.973 5.973 5.973 Fuel Energy (GWh) 23.893 19.808 17.781 Fuel Weight (tonne) 2134 1769 1588 Fuel Volume (gal) 627,292 520,035 466,822 Fuel Energy Savings (GW) n/a 4.085 6.112 Fuel Weight Savings (tonne) n/a 365 546 Fuel Volume Savings (gal) n/a 107,257 160,470

Table 10 – Operational Presence Fuel and Volume Savings

Gas Turbine Combined Cycle Combined Cycle

Only 1 2 Power Plant Efficiency (%) 25 35.3 45.4 Time at Sea (hr) 504 504 504 Max Electrical Energy (GWh) 16,128 16,128 16,128 Fuel Energy (GWh) 64.512 45.672 35.516 Fuel Weight (tonne) 5762 4079 3172 Fuel Volume (gal) 1,693,688 1,199,071 932,439 Fuel Energy Savings (GW) n/a 18.840 28.996 Fuel Weight Savings (tonne) n/a 1683 2590 Fuel Volume Savings (gal) n/a 494,617 761,249

Figure 14 – Percent fuel weight and volume reduction with increasing power plant efficiency

Figure 15 – Fuel weight savings with increasing efficiency

3.5 Summary

In this chapter, the major components of the system modeled in the previous chapter were analyzed for their effect on the weight of the ship. It was determined that analysis of weight of the electric generator, the auxiliaries and control components are beyond the scope of this work. However, algorithms for estimating the weight of the turbines, the heat exchangers, and the fuel were developed.

The turbine and heat exchanger analysis relied heavily on data available from commercially existing units of comparable size to those needed for ships operating in the Economical Transit mode of operation for Navy Ships. The fuel weight analysis used estimated efficiency gains to determine the weight and space savings that can be achieved by replacing a simple-cycle gas turbine with a combined cycle power plant.

In chapter 4, the methods developed in chapter 2 for analysis of efficiency will be exercised simultaneously with the weight analysis of this chapter to make recommendations for optimizing configurations of combined cycle power plants for all-electric ships.

CHAPTER 4 ANALYSIS RESULTS

In this chapter, implementation of an array of combined cycle power plant options will be considered as options for the replacement of the simple gas turbine standard used to power conventional navy ships. Doerry’s Economical Transit operational mode, which was introduced in Table 7, will be deployed for analysis. Recall that the maximum power required for Economical Transit mode is 22 MW. This reference will serve as the design target for the power output of all combined cycle power plant configurations discussed below. Using the tool developed in chapter 2, for evaluation of combined cycle power plants, and the weight analysis techniques presented in chapter 3, combined cycle power plant alternatives will be assessed for a variety of design scenarios. The nominal quality and pinch point values used in each of the section of this chapter are 95% and 20 °C, respectively, and the efficiency and exhaust gas temperature of the nominal gas turbine are 30% and 500 °C, respectively.

In section 4.1 the gross effects of recovering power from the gas turbine exhaust will be discussed. In section 4.2, the focus shifts to the impact that changing the quality of steam exiting the turbine at state 8 has on combined cycle power plants. Section 4.3 concentrates on the effects of varying the pinch point in the heat recovery steam generator (HRSG). Finally, section 4.4 utilizes the information of the previous three sections to assess the effects that altering gas turbine operating parameters (efficiency and exhaust gas temperature) have on the system. This analysis will help determine what type of gas turbine is best suited for retrofit with a combined cycle power plant. The chapter concludes with a brief summary of the findings in section 4.5.

Throughout the chapter, the primary assessment parameter employed in the evaluation of other system parameters is the power recovered from the gas turbine’s exhaust gases. This technique is intended to demonstrate that the power recovered from the exhaust gas stream is the driving factor that affects every parameter of the bottoming cycle power plant. However, the power recovered in the HRSG will not be displayed explicitly on the abscissa; rather, the temperature of the gas turbine exhaust exiting the HRSG (T5) will be displayed. This choice leverages the presumed proportional relationship between HRSG power and the temperature drop of the gas

turbine’s exhaust through that device. Recall from Eq. (8) that under steady state operation, if the mass flow rate and the specific heat of the gas turbine exhaust remain constant, then the power recovered in the HRSG is directly proportional to the temperature drop across the heat exchanger. Further, it is reasonable to assume that the gas turbine’s exhaust temperature at state 4 remains constant at 500 °C. Under these circumstances, the change in temperature at state 5 can

be used as a qualitative means for analyzing the power recovered by the HRSG. As T5 falls below 500 to 100 °C, the power recovered from the gas turbine exhaust increases from the lowest to the highest level considered in this work. Figure 16 highlights the measurement location of T5 in the combined cycle power plant.

Figure 16 –HRSG exhaust gas temperature is a qualitative measure recovered power

The tool, as described in the previous chapters, is limited to analysis for gas turbines, whose exhaust stream is directed through the HRSG of a simple, Rankine cycle steam turbine plant. Although the configuration analysis could be used to design a steam turbine cycle who’s HRSG absorbs power from the waste heat from any prime mover, the weight analysis provided in the previous chapter constricts the tool’s use to gas turbines. Moreover, only simple, Rankine cycle steam turbine power plants may be considered using this tool. Higher efficiency steam turbine

58 cycles that incorporate regeneration, re-heating, and other efficiency optimization methods are beyond the capability of this first approximation tool.

4.1 Case I – The Effects of Varying HRSG Exhaust Gas Temperature (T5)

The temperature drop of the gas turbine exhaust as it passes through the HRSG provides a good reference to qualitatively understand how the power recovered from the gas turbine exhaust affects various parameters of the combined cycle. The evolution of the combined cycle efficiency as a result of this power recovery is displayed in Figure 17. This graphic provides the framework for most of the figures that follow. The left axis is reserved for the parameter(s) considered later in the chapter, while the right axis is dedicated to tracking the combined cycle efficiency. The line representing the efficiency versus HRSG exhaust gas temperature is ubiquitous in the figures of this section.

Figure 17 – Efficiency versus T5 in the format used to evaluate other parameters

In most of the graphs in this section, a representation of the power recovered from the gas turbine exhaust will serve as the abscissa, the parameter under consideration will be the primary ordinate, and the efficiency of combined cycle power plant will frequently be represented by a 59

secondary ordinate located on the right side of the graph. This technique is used to underscore the importance of efficiency in the evaluation of combined cycle operation.

The variable parameters manipulated in this chapter are the steam quality of the bottoming cycle turbine exit, the pinch point in the heat recovery steam generator (HRSG), and the primary operational parameters of the gas turbine, exhaust gas temperature and operating efficiency. In this section, the values for steam turbine quality and HRSG pinch point are 95% and 20 °C, respectively, while the gas turbine modeled has an exhaust gas temperature of 500 °C and an efficiency of 30%. Doerry’s Economical Transit mode of operation provides the basis for evaluation.

The energy recovered from the gas turbine exhaust is used to drive the Rankine Cycle which produces work with the steam turbine. Figure 18 illustrates the effect that recovering varying amounts of power in the HRSG has on the power output of the steam turbine. The increased power output per unit of fuel consumed results in increased combined cycle efficiency as anticipated by Eq. (19). By augmenting the prime mover with a bottoming cycle, the efficiency of the combined cycle power plant can be elevated to as high as 42.4%. Initially, increasing power recovery in the HRSG corresponds with gains in both steam turbine output power and power plant efficiency. This trend continues until the HRSG exhaust gas exit temperature reaches 186 °C, when both the power output of the steam turbine and the efficiency of the combined cycle power plant drop precipitously. The power output of the steam turbine regresses from a maximum of 6.44 MW at 186 °C to 5.36 MW at 100 °C. Simultaneously, the combined cycle efficiency falls from 42.4% to just below 40%. If the earlier trends had continued, the 100 °C mark would have corresponded to the maximum steam turbine power output in excess of 7 MW, and combined cycle power plant efficiency greater than 45%.

The change in behavior in the region between 186 °C and 100 °C can be explained by the fact that the pinch point changed location. Recall the inequalities of Eqs. (13) and (14), as well as the discussions in sections 2.3.1 and 2.3.2 regarding the effects of moving the pinch point. In the region in which the steam turbine power and efficiency increase from 0 to 6.44 MW and from 30-42.4%, respectively, the pinch point is located at the interface between states 4 and 7, where

live steam exits the super heater. In this regime, power and efficiency increase with decreasing gas turbine exhaust temperature from the HRSG, as expected. However, as the exhaust gas

HRSG exit temperature (T5) drops, it brings T4b with it, eventually compelling the pinch point to move to the interface between states 4b and 6a where saturated liquid water enters the boiler, after being heated in the economizer. Figure 19 illustrates the change in pinch point location that is required to comply with the inequalities presented in Eqs. (13) and (14).

Figure 18 – The effects of changing T5 on the steam turbine power output

The change in pinch point position is followed by dramatic drops for both live steam (T7) and evaporator (T6a) temperature. Since any reduction in evaporator temperature results in a drop in HRSG pressure, the presumption of constant condenser pressure results in corresponding reductions in steam turbine pressure ratios in the regime between 186 °C and 100 °C. These drops in pressure ratio are manifested by the reduced efficiencies. The green line representing the live steam temperature (T7) in Figure 19 also indicates that although heat recovery from the exhaust gas stream continues to increase (T5 continues to decrease), when the pinch point moves to the interface between states 4b and 6a, less of that power is being used to elevate the temperature of the superheated steam.

In order to achieve the efficiency gains that are apparent in either pinch point regime, additional mechanical components must be added to the system. Using the analysis methods presented in chapter 3, the weights of the array of combined cycle configurations have been calculated. However, it is also apparent from the fuel weight analysis of chapter 3 that net fuel savings will coincide with the efficiency gains. The blue line in Figure 20 indicates the effect that varying the HRSG exhaust gas temperature has on the weight of the combined cycle power plant. The green line shows the response of the fuel to the various amounts of heat recovery in the HRSG, and the red line represents the effect on the mechanical components. Analysis of this figure shows that although the addition of mechanical components could result in a maximum weight increase of 128 tonne, the maximum fuel weight reduction achieved by elevating the efficiency of the power plant is approximately 400 tonne. The reduction in fuel weight dominates the weight impact that adding a bottoming cycle has on the combined cycle power plants proposed.

Figure 19 – The effect of reducing T5 on the location of the pinch point

Figure 21 illustrates the net effects that implementing the combined cycle has on the weight of the mechanical components, the fuel, and the net combined cycle system. This figure clearly illustrates that the reduction in fuel weight, represented by the green line, outpaces increase in

mechanical component weight, represented by the red line, for every power recovery level considered. The result, represented by the blue line, is an overall reduction in power plant weight, over the entire range considered. Note, however, that the rate of fuel savings slows dramatically at the 186 °C exhaust gas temperature mark. It should also be recognized that just before the pinch point is forced to relocate, the mechanical component’s weight contribution to the combined cycle is less than 75 tonne, but the fuel savings has reached its maximum of roughly 400 tonne.

Figure 20 – Gross effects on mechanical component, fuel, and net power plant weight

Figure 22 illustrates that while the gas turbine weight decreases with increased power recovery in the HRSG, the weight of the components that make up the Rankine Cycle bottoming plant increase and more than offset those savings. An interesting feature of Figure 22 is the domination of heat exchanger contribution to mechanical component weight. The HRSG, in particular grows from just 4 tonne for a nominal installation, to nearly 75 tonne at the pinch point transition. After the transition, the HRSG weight increases roughly linearly to nearly 95 tonne at 100 °C. Also, as

T5 approaches the 200 °C mark, the condenser undergoes the most dramatic rise in weight of any of the components in this plot. The nominal condenser’s weight contribution is roughly 10 tonne.

As power recovery from the exhaust gases increases, the condenser’s weight increases gradually to 11 tonne at the 186 °C exhaust gas temperature mark, but over the final 86 degrees temperature drop, the condenser weight increases rapidly to 38 tonne. The condenser’s considerable weight increase is directly related to the precipitous fall in live steam temperature, demonstrated by the green line in Figure 19. Because much of the heat recovered in the HRSG can no longer be converted into work, it simply passes through the turbine and where it must be transferred to the atmosphere through the condenser. This increase in waste energy transmitted through the bottoming cycle is absolutely counter-productive. Not only is the rate of non-useful energy conveyance through the Rankine cycle failing to produce work, but it also causes the area of condenser to increase in order to accommodate the increased heat rejection to atmosphere through the engineered system. Both activities are counter-productive to the goals of increasing power plant efficiency, and decreasing plant weight.

Figure 21 – The effects of varying T5 on the net weight

Figure 23 shows that as the temperature drop of the gas turbine exhaust gases increases across the HRSG progresses, initially, all of the heat exchanger surface areas are near zero, but as the temperature drop increases, the superheater and the evaporator increase in size fastest, and at comparable rates, until T5 falls to approximately 380 °C. At this point, the evaporator size begins

to increase dramatically to deal with the increased heat load, but the superheater temperature continues to increase in size at a relatively steady rate. Simultaneously, the economizer surface area begins to increase at a rate such that by the time the temperature in of the exhaust gases are lowered to 260 °C at the HRSG exit, the sizes of the economizer and superheater are equal, but growing at dramatically different rates. As the exhaust gases approach the pinch point transition, the condenser’s surface area remains fairly stable, but the superheater maintains its steady rate of increase to roughly 1535 m2. At this point, the economizer size has grown rapidly enough to nearly equal the ever-increasing evaporator, at 2945 and 3480 m2, respectively. After the pinch point transition, the required surface areas for all heat exchangers continue to increase to keep up with the increased heat load from the HRSG. The condenser begins to grow exponentially, while the evaporator appears to grow linearly, and the economizer and superheater appear to pass through relative maximum growth points and begin to level out or reduce in size.

Figure 22 – Mechanical components breakdown

Next, consider Figure 24, which illustrates the weight of the mechanical system components, the fuel, and the overall system for an Economical Transit-like mission, whose range is 1/4 that of the Economical Transit mode described by Doerry in Table 7. Since the power requirements of the ship are assumed to be the same the Economical Transit mode described by Doerry, the

weight and efficiency of the mechanical components are identical those in the analysis above. However, in this reduced-range configuration, the rate at which the mechanical components increase in weight outstrips the fuel weight savings at the point when the exhaust gas temperature drops to roughly 250 °C, well above the 186 °C pinch point transition. This illustration was presented to demonstrate the fact maximizing efficiency does not necessarily correlate to maximum weight savings. Both the power and the ship’s range must be taken into account to effectively minimize the weight of the ship.

Figure 23 – The effect of HRSG power variation on required heat exchanger surface areas

Since the effects of efficiency and weight are of paramount interest in this discussion, Figure 25 is presented; this plot compares those parameters directly. Recalling the analysis above, the unusual shape of this figure can be explained by recognizing that the two pinch point scenarios describe two substantively different combined cycle power plant configurations. When the pinch point lies at the interface between states 4 and 7 (as the power recovered by the HRSG increases), the weight of the mechanical components and the efficiency of the system increase, while the fuel weight decreases. On the contrary, when the pinch point lies at the interface between states 4b and 6a, as the power recovered by the HRSG increases, the efficiency of the

system decreases while the weight of the mechanical components and the fuel weight increase. The hook in the plot indicates that transition between the two major combined cycle power plant scenarios. The two sides of the hook could be considered two separate plots that converge at the transition point. These graphs illustrate the displacement of minimum weight from maximum efficiency introduced in the analysis of Figure 24. With the full range mission, the minimum weight and maximum efficiency coincide, but as the mission range is reduced, the minimum weight power plant moves to efficiency values that are not maximized.

Figure 24 – System weight for a 1/4 range Economical Transit-type mission

Finally, Figure 26 presents the weight impact versus efficiency information of Figure 25, as a net savings. The effects are identical to those discussed above with minimum weight power plant configurations having lower efficiency than power plants optimized for efficiency, with the effect being exaggerated for reduced-range power plants. In subsequent sections, the pinch point in the heat recovery steam generator (HRSG), and the operational parameters of exhaust gas temperature, and operating efficiency for the gas turbine will be varied to demonstrate the effects of increasing or decreasing those parameters on the efficiency, the steam turbine power, the net

weight differentials, and other parameters that contribute heavily to this combined cycle power plant analysis.

Figure 25 – Net weight versus for full, 1/2, 1/3, and 1/4 Economical Transit trip durations

4.2 Case II – The Effects of Varying Steam Quality (x8)

In this section, Doerry’s Economical Transit mode remains the basis for evaluation, and all of the parameters of the previous section’s combined cycle analysis are left unchanged, except the quality of the steam turbine’s exhaust. Qualities of 89%, 100%, and 104% are compared to the nominal results of the previous section, whose quality was 95%.

Figure 27 and Figure 28 illustrate the similar effects that reducing the steam turbines exhaust’s quality has on the efficiency and the steam turbine power generation, respectively. The configuration with maximums for both efficiency and steam power generation is the nominal design, while the power plant whose steam exhausts the turbine at 104% quality generates the lowest potential for both efficiency and steam turbine power generation gains.

Figure 26 – Net weight reduction versus efficiency for several Economical Transit-type trips

The extreme cases achieve efficiencies of 42.4% and 41.4%, and produce 6.436 and 6.090 MW of steam turbine power, respectively. The difference between the maximum for these configurations values is separated by just 346 kW, or 5.7%, of additional steam turbine power generation for the nominal configuration. The two remaining combined cycle plants are capable of achieving maximum efficiencies and steam turbine power outputs of 42.3% and 6.406 MW for the 89% exhaust quality variation and 42.1% and 6.335 MW for the configuration with saturated exhaust.

Figure 29 shows that lower quality combined cycle power plant designs shed weight earlier in the heat recovery process, but the nominal design ultimately generates the most weight savings. It is interesting that in the three figures presented so far in this chapter note in, the efficiency, power, and weight converge as T5 approaches 100 °C for all value of quality. Figure 30 shows that changing the quality has very little effect on the weight savings per day.

Figure 27 – Effects of quality on combined cycle efficiency

Figure 28 – Effects of quality on combined steam turbine power

Recalling the dominance of the heat exchangers in the mechanical components weight breakdown of Figure 22, an interesting phenomenon occurs when considering the effect of steam turbine exhaust’s quality on the weight of the HRSG. Figure 31 shows that although the condenser weights vary very little with the quality, for any of the configurations, the HRSG weights differ significantly. Predictably, since lower quality models gain efficiency faster, they are generally lighter, but the 89% steam quality exhaust configuration loses its weight dominance

noticeably when T5 approaches 100°C. This effect can be explained by considering the presumed weight dominance of the economizer and the evaporator demonstrated in Figure 23.

Figure 29 – Effects of quality on weight savings

Figure 32 demonstrates that the weight of the evaporator begins to fall around 115 °C for all of the configuration variations considered in this section. For the higher qualities, the evaporator area far exceeds the economizer area, but for the lowest quality, the economizer area is on par with that of the evaporator, so the drop in area affects the overall weight of the HRSG.

Figure 30 – Effects of quality on weight impact per day for full range mission

Figure 31 – Effects of quality on HRSG and condenser weight

While this finding is interesting, it should be recalled from the analysis of section 4.1, that the mechanical weight phenomenon described above is insignificant to the overall combined power plant design for two reasons: first, the fuel weight dominance outstrips these meager savings, and second, since this effect occurs after the pinch point transition, it is in the region where weight is increasing. Without the benefits of increased power production and efficiency, it is very unlikely that a design in this region would be considered for implementation.

Figure 32 – Effects of quality on mechanical component weight

4.3 Case III – The Effects of Varying HRSG Pinch Point

Heat exchanger analysis is omitted in this section and the next, because of the relative insignificance of the mechanical component weight on the overall system. Notwithstanding this shift, a similar approach to that of the previous section is employed here. Two additional pinch point configurations 1 °C and 40 °C have been added to the information generated for a 20 °C pinch point in section 4.1 for comparative analysis. Of course the quality is reset to 95%, but the

remaining combined cycle parameters are left unchanged, and Doerry’s Economical Transit mode remains the basis for evaluation.

Figure 33 and Figure 34 show very little difference between the three pinch point configurations as the heat energy recovery increases, until the transition point is reached. For the 40 °C variation, this transition occurs at 200 °C, markedly earlier in the heat recovery progression than the 186 °C transition point observed for the nominal plant design. It is also noteworthy that the higher pinch point allowance results in reductions in both maximum efficiency and steam turbine power, from 4.24% and 6.439 MW for the nominal installation, to 41.3% and 6.035 MW in the 40 °C pinch point configuration.

Figure 33 – Effects of changing the HRSG pinch point

The trend of higher efficiency and steam turbine power generation at lower T5 pinch point transition values for smaller pinch points extends to the 1 °C pinch point configuration, whose efficiency and steam turbine power generation levels are elevated to 43.5% and 6.814 MW, at 160 °C. Nearly 800 kW of additional power can be generated if the pinch point is minimized compared to the nominal 20 °C pinch point results. That represents roughly a 13% increase in power production from the steam turbine.

Figure 34 – Effects of changing the HRSG pinch point on steam turbine power production

The impact on net weight savings is not as significant as the power production and efficiency gains. However, the smallest pinch point configuration remains most beneficial. The net weight savings are 330, 329, and 321 tonne respectively for 1 °C, 20 °C, and 40 °C pinch point configurations respectively. The weight analysis findings are presented graphically in Figure 35.

Interestingly, Figure 36 shows that prior to the pinch point transition; there is very little difference in the net weight savings per day for any of the pinch point design variations. Significant weight differences are observed after each respective configuration goes through its pinch point transition, where the behaviors of substantially decreased weight savings with increased heat recovery return in a manner analogous to the behavior described in section 4.1. The explanation of that behavior will not be expanded upon here.

Figure 35 – Effects of pinch point variation on net weight savings

Figure 36 – Effects of pinch point variation on net weight savings per day

4.4 Case IV – The Effects of Gas Turbine Performance

This section provides the most stark contrast between varied parameters. The exhaust steam quality, and the pinch point are returned to their nominal levels of 95% and 20 °C, but the gas turbine operating parameters are manipulated.

The nominal gas turbine of section 4.1 operated with an efficiency of 30%, and produced exhaust gas at 500 °C, which was supplied to the HRSG. This is very typical for gas turbines used for power production. Gas turbines with higher efficiency produce exhaust gases with lower temperature. In this case, “the high efficiency gas turbine” operates at 35% efficiency, and produces exhaust gas at 450 °C. Similarly, gas turbines with lower efficiency produce exhaust gases with higher temperature. The “low efficiency gas turbine” presented in this chapter operates at 25% efficiency, and produces exhaust gas at 550 °C.

Figure 37 illustrates the efficiency gains achievable by adding a combined cycle power plant to each of the three gas turbines described above. Since the pinch point and steam turbine exhaust qualities are assumed equal, the pinch point transition points are identical, and match the nominal gas turbine at 186 °C. Also, it is clear that the high efficiency turbine’s maximum efficiency is 5.1% higher than that of the low efficiency gas turbine, but this does not tell the whole story of efficiency impact for various gas turbines.

Since efficiency gains come from increased steam power output, the low efficiency gas turbine prime mover produces work from recovered power, while the high efficiency turbine produces the least power. Figure 38 illustrates this point clearly. Recall that the nominal configuration produced 6.439 MW of power. Similarly, the maximum power output of the bottoming cycle falls from 8.235 MW for the low efficiency gas turbine prime mover combined cycle power plant to 4.940 MW for the high efficiency version. Since the lower efficiency gas turbine installations produce higher temperature exhaust gases, they are able to achieve greater net efficiency gains before reaching the pinch point transition.

Figure 37 – Efficiency response to various gas turbine prime movers

Figure 38 – Steam turbine power output for various gas turbine prime mover configurations

For example, the low efficiency turbine increases its efficiency from 25% to 40.0%, and thus experiences a 60% increase in overall power plant efficiency at the pinch point transition. The nominal gas turbine design is produces efficiency gains from 30 to 42.4% at the pinch point transition, a 41% increase, and the high efficiency gas turbine elevates its efficiency from 35 to 45.1% a the pinch point transition, an increase of just 28%.

Predictably, Figure 39 and Figure 40 show that the weight savings for a low efficiency gas turbine- powered combined cycle can save a net of 528 tonne for a full-range Economical Transit monde mission, which translates to 60.4 tonne per day over the course of the mission. These savings dwarf comparable savings of 329 tonne and 37.6 tonne per day for the nominal configuration, and 204 tonne and 23.3 tonne per day for the high efficiency prime mover installations. Clearly, greater benefit is derived from adding a combined cycle to a less efficient gas turbine that produces higher exhaust gas temperatures to feed the HRSG.

Figure 39 – Net weight savings for various gas turbine prime mover configurations

Figure 40 – Net weight savings per day for various prime mover configurations

4.5 Analysis Summary

In this chapter, the tool developed in chapter 2 for configuration analysis were married with the weight analysis techniques of chapter 3 to explore the effects of varying several parameters in a 22 MW combined cycle power plant.

The presumption from chapter 3 that for large marine applications, the weight penalty associated with adding equipment to implement a combined cycle power plant is more than offset by the weight savings achieved by increasing the efficiency of the overall power plant was borne out in every section of this chapter’s analysis. Another interesting finding is that as the range of the mission for ship-deployment decreases, the point of optimal weight separates from the optimal efficiency point. This observation is an artifact of the reduced dominance of fuel weight savings over mechanical systems weight gains when a bottoming cycle power plant is used to augment the performance of ship’s power plant. This phenomenon was discovered in section 4.1.

The effects of changing the pinch point from the interface between T4 and T7 to the interface between T4b and T6a in the HRSG can be observed in every figure of this chapter. The analysis shows that the interface between T4 and T7 exhibits the desired effects of increasing efficiency and reducing weight with increased power recovery from the gas turbine’s exhaust. On the contrary, when the pinch point is located at the interface between T4b and T6a, the weight and efficiency trend down with increased heat recovery.

It has also been established that the implementation of a combined cycle is most effectual when applied to a low efficiency gas turbine, as described in section 4.4. Although the maximum efficiency of the high-efficiency turbine is noticeably greater than that of the low-efficiency turbine, the weight savings of the low-efficiency model severely outpaces the high-efficiency option.

Some of the less impactful findings of the chapter were discovery that the heat exchangers dominate the mechanical weight component’s effects. This discovery was made in section 4.1, and underscored in section 4.2. The analysis of section 4.2 further revealed that the efficiency gains, and subsequently weight reduction can be achieved with slightly less heat recovery by the designing the bottoming cycle steam turbine to exhaust lower-quality steam; however “medium- quality” exhaust steam was found to produce maximum efficiency. In any case, the maximum efficiencies achievable were in a similar range. In section 4.3, varying the pinch point was found to have very little effect on the behavior of the combined cycle power plant other than allowing lowering the maximum amount of energy recovered, and subsequently the efficiency gains with higher pinch point designs.

CHAPTER 5 CONCLUSIONS & FUTURE WORK

“Going full electric was/is the same step as going from sail to steam,” according to Commander Ian Atkins, Royal Navy Assistant Attaché at the British Embassy in Washington. As the U.S. Navy follows the lead of its British allies, comprehensive tools for analyzing the performance of the power plant options available is a necessity [2]. This work has been an exercise in the development of a modeling tool for analyzing configuration and weight of combined gas turbine and steam cycle systems. Close attention was paid to the impact of weight, because of the intended use in marine applications. In chapter 2, the components and modeling techniques for the topping and bottoming cycles were presented, in an effort to describe the solution methodology used for the analysis of chapter 4.

In chapter 3, the major components of the system were analyzed for their effect on the weight of the ship. It was determined that analysis of weight of the electric generator, the auxiliaries and control components are beyond the scope of this work. However, algorithms for estimating the weight of the turbines, the heat exchangers, and the fuel were developed. It was estimated that adding weight to the power plant in the form of steam cycle components would be more than offset by the downsizing of the gas turbine prime mover and, more importantly, reducing the fuel weight by increasing the overall efficiency of the power plant. Commercially available hardware and gas turbine scaling techniques were used to confirm these assertions. For example, the turbine and heat exchanger analysis relied heavily on data available from existing commercial units.

The tool developed from the analyses in chapters 2 and 3 were used to estimate the efficiency gains and determine the weight impact of replacing a simple-cycle gas turbine with a combined cycle power plant in chapter 4. The presumption from the increase in weight penalty associated with adding equipment would be offset by the weight savings achieved by reducing fuel weight was borne out in the analysis. It was also established that the implementation of a combined cycle is most effectual when applied to a low efficiency gas turbine.

The analysis also revealed that the point of optimal weight does not always coincide with the point of optimal efficiency. Moreover, the effects of changing the pinch point location were

observed, and the analysis showed that the interface between T4 and T7 exhibits the most desirable affects for implementation on an actual ship.

From the analysis, it can be concluded that the ideal gas turbine for retrofitting with a combined cycle power plant would be one with a relatively low-efficiency gas turbine. The design should be made such that the pinch point lies at the interface of the HRSG where the high temperature exhaust gas completes the superheating of the steam, just before it enters the steam turbine.

However, if maximum efficiency were the desired effect in combined cycle design, a high- efficiency gas turbine should be selected, the pinch point in the HRSG should be minimized, and some further analysis of the steam turbine exhaust could be performed in order to pin-point the most ideal steam turbine exhaust quality.

Similar work could be performed, developing tools for all of the ship architectures described in section 1.2. Additional work that could be commissioned for analysis in the all-electric ship implementation is the ideal configuration for converting the mechanical shaft power available at the steam and gas turbines into electrical energy for distribution throughout the vessel. The conventional method for conversion is to decrease the speed of the shaft through a mechanical gearbox, then use a generator to produces alternating current electricity with a desired frequency. This conventional method requires additional auxiliary systems, for safe operation. All of the additional components need to be considered in a separate weight analysis. However, there are alternatives, such as implementation of high speed generators that produce power in a much smaller form factor than conventional generators. One drawback to this method is that the electrical frequency of such machines is not compatible with most electrical devices, so power electronic converters would be required to produce electricity with the characteristics required for use in the wide variety of electricity-consuming devices onboard a ship. DC zonal distribution should not be excluded from the options considered for ship power architectures.

There are also a variety of additional thermal and mechanical considerations that should be considered, such as heating/cooling integration, compact heat exchangers that reduce size, while maintain performance, consideration of water piping, and air ducting networks for minimizing weight, and thermal anticipation strategies of the improvement of the thermal system’s response to the intermittent use of auxiliary systems, and anticipation of future loads. Implicit in the need for a robust electrical system is the need for an equally robust cooling system. The US Department of Defense has explicitly stated that heat is the second leading cause of failure for electrical equipment. The US Department of Defense has explicitly stated that heat is the second leading cause of failure for electrical equipment [5].

BIBLIOGRAPHY

[1] T. J. McCoy, "Trends in Ship Propulsion," in IEEE Xplore, Chicago, 2002.

[2] E. H. Lundquist, "The All-Electric Ship: Ready to Take Over?," Naval Forces, vol. 32, no. 1, pp. 41-45, 2011.

[3] C. N. H. Doerry, "Sizing Power Generation and Fuel Capacity of the All Electric Warship," IEEE, 2007.

[4] N. L. Doerry, H. U. Robey, J. U. Amy and C. Petry, "Powering the Future with the Integrated Power System," Naval Engineers Journal, no. May, pp. 267-279, 1996.

[5] C. R. Holsonback and T. M. Kiehne, "Thermal Aspects of a Shipboard Integrated Electric Power System," Office of Naval Research, 2010.

[6] J. Fernandez-Seara, A. Vales and V. Manuel, "Heat Recovery System to Power an Onboard NH3-H2O Absorption Refrigeration Plant in Trawling Chiller Fishing Vessels," Applied Thermal Engineering, vol. 18, pp. 1189-1205, 1998.

[7] F. Haglind, "A Review on the Use of Gas and Steam Turbine Combined Cycles as Prime Movers for Large Ships. Part I: Background and Design," Energy Conversion and Management, vol. 49, pp. 3458-3467, 2008.

[8] F. Haglind, "A Review on the Use of Gas and Steam Turbine Combined Cycles as Prime Movers for Large Shipe. Part II: Previous Work and Implications," Energy Conversion & Management, vol. 49, pp. 3468-3475, 2008.

[9] F. Haglind, "A Review on the Use of Gas and Steam Turbine Combined Cycles as Prime Movers for Large Ships. Part III: Fuels and Emissions," Energy Conversion & Management, vol. 49, pp. 3476-3482, 2008.

[10] L. O. Nord and O. Bolland, "Designs and Off-Design Simulations of Combined Cycles for Offshore Oil and Gas Installations," Applied Thermal Engineering, vol. 54, pp. 85-91, 2013.

[11] S. R. C. Young, J. R. C. Newell and G. R. C. Little, "Beyond Electric Ship," Naval Engineers Journal, vol. 113, no. 4, pp. 79-98, 2001.

[12] I. Emmanuel-Douglass, "Performance Evaluation of Combined Cycles for Cruise Ship Applications," in IMECE2008-67393, Boston, MA, 2008.

[13] F. Haglind, "Variable geometry gas turbines for improving the part-load performance of marine combined cycles - Combined cycle performance," Applied Thermal Engineering, vol. 31, pp. 467-476, 2011.

[14] M. Dzida, J. Girtler and S. Dxida, "On the possible increasing of efficiency of ship power plant with the system combined of marine Diesel engine, gas turbine and steam turbine in case of main engine cooperation with the gas turbine fed in series and the steam turbine," Polish Maritime Research, vol. 16, pp. 26-31, 2009.

[15] G. Van Wylen, R. E. Sonntag and C. Borgnakke, Fundamentals of Classical Thermodynamics, 4th Edition, New York, New York: John Wiley & Sons, 1994.

[16] D. Brandt and R. Wesorick, "Gas Turbine Design Philosophy," GE Industrial & Power Systems, Schanectady, NY.

[17] F. J. Brooks, "GE Gas Turbine Performance Characteristics," GE Power Systems, Schanectady, NY, 2004.

[18] E. B. &. L. H. B. Woodruff, "Steam Plant Operation," McGraw-Hill, Inc., New York, NY, 1977.

[19] K. &. T. E. Rawson, Basic Ship Theory, Fifth ed., vol. 2, Woburn, Massachussetts: Butterworth Heinemann, 2001.

[20] F. P. Incoproper and D. P. DeWitt, Fundamentals of Heat and Mass Transfer, 4th Edition, New York, New York: John Wiley & Sons, 1996.

BIOGRAPHICAL SKETCH

Michael Coleman is the second of two sons born to Norwood Sr. and Alice Coleman in Wilmington, DE. Both parents were first generation college graduates at Delaware State University, who led successful careers in the human services fields. With their support and encouragement, Michael matriculated through St. Edmunds and Archmere Academies, while attending the FAME (Forum for the Advancement of Minorities in Engineering) in Delaware.

After graduation from High School, Michael attended Florida A&M University, and acquired a Bachelor’s degree in Mechanical Engineering. Following 3-years of employment at the Torrington Bearings Plant in Cairo, GA, Michael began working for the Center for Advanced Power Systems at Florida State University, where he works with the facilities group, assisting procurement, installation, maintenance, data acquisition and machine control programming, associated with the testing of machinery up to 5 MW level.

While working at the University, Michael took advantage of the benefit offered to employees to advance his education. Michael’s interest in Thermodynamics led him to Dr. Juan Ordonez, who advised him in his quest for a deeper understanding of engineering concepts.