HVDC Transmission PDF

High Voltage Direct Current Transmission – Proven Technology for Power Exchange 2 Contents

Chapter Theme Page

Contents 3 1Why High Voltage Direct Current? 4 2 Main Types of HVDC Schemes 6 3 Converter Theory 8 4Principle Arrangement of an 11 HVDC Transmission Project 5 Main Components 14 5.1 Thyristor Valves 15 5.2 Converter Transformer 18 5.3 Smoothing Reactor 21 5.4 Harmonic Filters22 5.4.1 AC Harmonic Filter 23 5.4.2 DC Harmonic Filter 25 5.4.3 Active Harmonic Filter 26 5.5 Surge Arrester 28 5.6 DC Transmission Circuit 5.6.1 DC Transmission Line 31 5.6.2 DC Cable 33 5.6.3 High Speed DC Switches 34 5.6.4 Earth Electrode 36 5.7 Control & Protection 38 6System Studies, Digital Models, 45 Design Specifications 7Project Management 46

3 1 Why High Voltage Direct Current ?

1.1 Highlights from the High Line-Commutated Current Sourced Self-Commutated Voltage Sourced Voltage Direct Current (HVDC) History Converters Converters The transmission and distribution of The invention of mercury arc rectifiers in Voltage sourced converters require electrical energy started with direct the nineteen-thirties made the design of semiconductor devices with turn-off current. In 1882, a 50-km-long 2-kV DC line-commutated current sourced capability. The development of Insulated transmission line was built between converters possible. Gate Bipolar Transistors (IGBT) with high Miesbach and Munich in Germany. voltage ratings have accelerated the At that time, conversion between In 1941, the first contract for a commer- development of voltage sourced reasonable consumer voltages and cial HVDC system was signed in converters for HVDC applications in the higher DC transmission voltages could Germany: 60 MW were to be supplied lower power range. only be realized by means of rotating to the city of Berlin via an underground DC machines. cable of 115 km length. The system The main characteristics of the voltage with ±200 kV and 150 A was ready for sourced converters are a compact design, In an AC system, voltage conversion is energizing in 1945. It was never put four-quadrant operation capability and simple. An AC transformer allows high into operation. high losses. power levels and high insulation levels within one unit, and has low losses. It is Since then, several large HVDC systems Siemens is offering voltage sourced a relatively simple device, which requires have been realized with mercury arc converters for HVDC applications with valves. ratings up to 250 MW under the trade little maintenance. Further, a three-phase plus synchronous generator is superior to a name HVDC Power Link Universal The replacement of mercury arc valves Systems. DC generator in every respect. For these by thyristor valves was the next major reasons, AC technology was introduced development. The first thyristor valves This paper focuses upon HVDC trans- at a very early stage in the development were put into operation in the late mission systems with high ratings, i.e. of electrical power systems. It was soon nineteen-seventies. with line-commutated current sourced accepted as the only feasible technology converters. for generation, transmission and distri- The outdoor valves for Cahora Bassa bution of electrical energy. were designed with oil-immersed thyristors with parallel/series connection However, high-voltage AC transmission of thyristors and an electromagnetic firing links have disadvantages, which may system. compel a change to DC technology: Further development went via air- • Inductive and capacitive elements of insulated air-cooled valves to the air- overhead lines and cables put limits insulated water-cooled design, which is to the transmission capacity and the still state of the art in HVDC valve design. transmission distance of AC transmission links. The development of thyristors with higher •This limitation is of particular signi- current and voltage ratings has eliminated ficance for cables. Depending on the the need for parallel connection and required transmission capacity, the reduced the number of series-connected system frequency and the loss eva- thyristors per valve. The development of luation, the achievable transmission light-triggered thyristors has further distance for an AC cable will be in the reduced the overall number of range of 40 to 100 km. It will mainly components and thus contributed to be limited by the charging current. increased reliability. • Direct connection between two AC Innovations in almost every other area systems with different frequencies is of HVDC have been constantly adding not possible. to the reliability of this technology with • Direct connection between two AC economic benefits for users throughout systems with the same frequency or the world. a new connection within a meshed grid may be impossible because of system instability, too high short-circuit levels or undesirable power flow scenarios. HVDC = high voltage direct current DC = direct current Engineers were therefore engaged over AC = alternating current generations in the development of a IGBT = insulated gate bipolar technology for DC transmissions as a transistor supplement to the AC transmissions.

4 1.2 Te chnical Merits of HVDC 1.3 Economic Considerations 1.4 Environmental Issues The advantages of a DC link over an AC For a given transmission task, feasibility An HVDC transmission system is basi- link are: studies are carried out before the final cally environment-friendly because decision on implementation of an HVAC improved energy transmission possi- •A DC link allows power transmission or HVDC system can be taken. Fig.1-1 bilities contribute to a more efficient between AC networks with different shows a typical cost comparison curve utilization of existing power plants. frequencies or networks, which can between AC and DC transmission not be synchronized, for other reasons. considering: The land coverage and the associated • Inductive and capacitive parameters right-of-way cost for an HVDC overhead do not limit the transmission capacity •AC vs. DC station terminal costs transmission line is not as high as that or the maximum length of a DC •AC vs. DC line costs of an AC line. This reduces the visual overhead line or cable. The conductor •AC vs. DC capitalised value of losses impact and saves land compensation for cross section is fully utilized because new projects. It is also possible to in- there is no skin effect. The DC curve is not as steep as the AC crease the power transmission capacity curve because of considerably lower line for existing rights of way. A comparison For a long cable connection, e.g. beyond costs per kilometre. For long AC lines between a DC and an AC overhead line 40 km, HVDC will in most cases offer the cost of intermediate reactive power is shown in Fig. 1-2. the only technical solution because of compensation has to be taken into the high charging current of an AC cable. account. This is of particular interest for transmission across open sea or into large The break-even distance is in the range cities where a DC cable may provide the of 500 to 800 km depending on a number only possible solution. of other factors, like country-specific cost elements, interest rates for project •A digital control system provides financing, loss evaluation, cost of right accurate and fast control of the active of way etc. power flow. •Fast modulation of DC transmission AC- DC- power can be used to damp power tower tower oscillations in an AC grid and thus improve the system stability. Fig. 1-2: Typical transmission line structures for approx. 1000 MW

There are, however, some environmental issues which must be considered for the Costs Total AC Cost converter stations. The most important ones are: •Audible noise Total •Visual impact DC Cost • Electromagnetic compatibility • Use of ground or sea return path in monopolar operation AC Losses DC Losses In general, it can be said that an HVDC system is highly compatible with any environment and can be integrated into it without the need to compromise on DC Line any environmentally important issues of today. AC Line DC Terminals

AC Terminals

Break-Even Transmission Fig. 1-1: Distance Distance Total cost/distance

5 2 Main Types of HVDC Schemes

2.1 DC Circuit 2.2 Back-to-Back Converters 2.3 Monopolar Long-Distance Tr ansmissions The main types of HVDC converters are The expression Back-to-back indicates distinguished by their DC circuit arrange- that the rectifier and inverter are located For very long distances and in particular ments. The following equivalent circuit in the same station. for very long sea cable transmissions, a is a simplified representation of the return path with ground/sea electrodes DC circuit of an HVDC pole. Back-to-back converters are mainly used will be the most feasible solution. for power transmission between adjacent AC grids which can not be synchronized. They can also be used within a meshed grid in order to achieve a defined power HVDC ±±Id flow. Cable/OHL

Ud1 Ud2

System 1 System Electrodes 2 System AC AC HVDC Fig. 2-1: Equivalent DC circuit Fig. 2-4: Monopole with ground return path The current, and thus the power flow, is System 2 System controlled by means of the difference 1 System In many cases, existing infrastructure or AC between the controlled voltages. The AC environmental constraints prevent the current direction is fixed and the power use of electrodes. In such cases, a direction is controlled by means of the metallic return path is used in spite of voltage polarity. The converter is de- Fig. 2-2: Back-to-back converter increased cost and losses. scribed in the next section.

HVDC Cable/OHL System 1 System LVDC 2 System AC AC

Fig. 2-5: Monopole with metallic return path

HVDC = high voltage direct current DC = direct current AC = alternating current Ud = DC voltage 12-pulse Fig. 2-3: Back-to-back converter Id = DC current Station Vienna Southeast OHL = overhead line LVDC = low voltage direct current

6 2.4 Bipolar Long-Distance 2.4.1 Bipole with Ground Return 2.4.2 Bipole with Dedicated Metallic Tr ansmissions Path Return Path for Monopolar Operation A bipole is a combination of two poles This is a commonly used configuration If there are restrictions even to temporary in such a way that a common low voltage for a bipolar transmission system. The use of electrodes, or if the transmission return path, if available, will only carry a solution provides a high degree of distance is relatively short, a dedicated small unbalance current during normal flexibility with respect to operation with LVDC metallic return conductor can be operation. reduced capacity during contingencies considered as an alternative to a ground or maintenance. return path with electrodes. This configuration is used if the required transmission capacity exceeds that of a single pole. It is also used if requirement to higher energy availability or lower load HVDC HVDC rejection power makes it necessary to Cable/OHL Cable/OHL split the capacity on two poles. Electrodes LVDC During maintenance or outages of one Cable/OHL System 2 System System 1 System System 2 System pole, it is still possible to transmit part 1 System AC AC AC of the power. More than 50% of the HVDC AC HVDC transmission capacity can be utilized, Cable/OHL Cable/OHL limited by the actual overload capacity of the remaining pole. Fig. 2-6: in bipolar balanced operation (normal) Fig. 2-9: in bipolar balanced operation (normal) The advantages of a bipolar solution over a solution with two monopoles are Upon a single-pole fault, the current of reduced cost due to one common or no the sound pole will be taken over by the 2.4.3 Bipole without Dedicated return path and lower losses. The main ground return path and the faulty pole Return Path for Monopolar Operation disadvantage is that unavailability of the will be isolated. return path with adjacent components A scheme without electrodes or a will affect both poles. dedicated metallic return path for monopolar operation will give the lowest HVDC initial cost. Cable/OHL Electrodes HVDC System 2 System System 1 System Cable/OHL AC AC HVDC Cable/OHL System 2 System System 1 System Fig. 2-7: in monopolar ground return AC AC operation (converter pole or OHL outage) HVDC Cable/OHL Following a pole outage caused by the converter, the current can be commutated Fig. 2-10: in bipolar balanced operation (normal) from the ground return path into a metallic return path provided by the Monopolar operation is possible by HVDC conductor of the faulty pole. means of bypass switches during a converter pole outage, but not during an HVDC conductor outage. HVDC Cable/OHL A short bipolar outage will follow a converter pole outage before the bypass Electrodes operation can be established. System 2 System System 1 System AC AC HVDC Cable/OHL

Fig. 2-8: in monopolar metallic return operation (converter pole outage)

7 3 Converter Theory

3.1 Bridge Circuit Function The angle between the time at which 3.2 12-Pulse Group and Converter the valve voltage becomes positive and Tr ansformer Current flows through the valves when the firing time (start of commutation) is the voltage between the anode and referred to as the firing delay. Fig. 3-2 HVDC converters are usually built as 12- cathode is positive. For the valve to pulse circuits. This is a serial connection shows that for a firing delay of 90°, the commutate the current, there must be average voltage equals zero. i.e. the of two fully controlled 6-pulse converter a positive potential (voltage), and the positive and negative areas of the curve bridges and requires two 3-phase sys- thyristor must have firing pulses. In the – voltage against time – cancel each tems which are spaced apart from each reverse direction, i.e. when the potential other out. No active power flows through other by 30 electrical degrees. The phase between anode and cathode is negative, the converter. difference effected to cancel out the a firing pulse has no effect. The flow of 6-pulse harmonics on the AC and DC current in a valve ends when the voltage When the firing delay is greater than side. between anode and cathode becomes 90°, the negative voltage/time areas negative. The instant when current begins dominate, and the polarity of the average to flow through a valve, or to commutate direct voltage changes. Due to physical from one valve to another, can be delayed reasons, the direction of the current does by postponing the firing. This method not change. (The thyristor valves conduct 1 3 2 permits the average value of the outgoing current only in one direction.) When the voltage of the rectifier to be changed. direction of energy flow is reversed, the The firing pulses are generated by syn- delivery changes to the supply side. The chronizing the network using an elec- rectifier becomes an inverter which tronic control device. These pulses can be delivers energy to the AC network. displaced from their ”natural firing“ point, 4 which is the point where the two phase The average value of the direct voltage voltages intersect. The method of firing- as a function of the firing delay is given by: 1Valve Branch pulse displacement is called phase 2 Double Valve Udiα = 1.35 * UL * cos α control. 3 Valve Tower UL = secondary side line voltage 4 6-pulse Bridge α = firing angle γ = extinction angle Fig. 3-3: Arrangement of the valve branches in a 12-pulse bridge DC current in each valve and phase 1351 0° L1 L2 L3 Udi 60° t 0° 120° Ud ω α = 0°

62 4 62 0° Id

i2 t ω α = 60° i3 60° i4

i5 i ω t α = 90° 6 90 ° γ = 90° i L1 i L2 150° ω t iL 3 α = 150° γ = 30° Id 153

L1 180° ω t L2 Ud α = 180° γ = 0° L3

4 6 2 Fig. 3-1: Six-pulse converter bridge Fig. 3-2: DC voltage of bridge converter 8 as a function of α 3.3 Reactive Power as a Function α = 60° of Load The curve of reactive power demand of an HVDC station with changing active power P can be calculated from equation: Udi α Q = P * tan [ arc cos ( cos α - dx)] In Fig. 3-5, the reactive power demand of a converter is presented under three ω t different control methods.

If the terminal DC voltage Ud and the firing angle α (or the extinction angle γ of an inverter) are held constant, curve (1) will be obtained. If, however, Uv is Secondary Voltage held constant (Udi = const regulation), a of the Transformer linear curve such as (2) is obtained. The Basic AC Current power of a converter can also be changed α when the (nominal) current is held constant by varying the DC voltage. Curve (3) shows the reactive power demand for this control method. It is important to note that the entire area between curves ϕ (1) and (3) is available for reactive power control. Each point within this area can be set by the selection of firing angles α and ß (or γ).

Fig. 3-4: Current displacement with angle control

Q/PN HVDC DC Circuit 1. 2 UdN = PdN Rec/IdN 3 1. 0 UdN => nominal DC voltage 12-pulse

IdN => nominal DC current 0.8 P => nominal DC active power dNRec 0.6 at the rectifier

0.4 2

0.2 1 P/PN

0.5 1. 0

1 Ud = const; α = const ( = const) 2 Ud = const; Uv = const 3 Id = const

Fig. 3-5: Reactive power demand of an HVDC converter

dx =relative inductive voltage drop Uv =valve voltage Ud = DC voltage 12-pulse α = firing angle ß = 180°-α γ =extinction angle

9 3 Converter Theory

3.4 Reactive Power Control 420 kV 50 Hz 420 kV 50 Hz The possibility of electronic reactive power control as demonstrated in the preceding section is used only to a very limited degree in HVDC technology. This is due to economic reasons. Both control reactive power and commutation reactive power are increased by the reduction of the DC voltage and the corresponding increase of current. However, load losses increase with the square of the current. For this reason, application is limited to Q = 103 Mvar Q = 103 Mvar the light loads where the necessary filter circuits produce a considerable overcom- Q = 103 Mvar Q = 103 Mvar pensation for the reactive power required by the converter. Fig. 3-6 depicts the reactive power control Q = 103 Mvar Q = 103 Mvar of the Dürnrohr HVDC link. In this system, Fig. 3-6: Reactive-power compensation and a compensation to ± 60 Mvar was spe- control of an HVDC back-to-back link cified. Compliance with the Q limit is achieved by load-dependent switching Q (Mvar) of a capacitor bank and one of the two 100 high-pass filters. Electronic reactive power is used only in the light load range. 80 Normally, there is a difference between the connect and disconnect points of the reactive power elements. This pro- 60 vides a ”switching hysteresis” which Over load Over

prevents too many switching operations Normal load or even a ”pumping”. 40 educed minimum load

20 Normal minimum load R

100 200 300 400 500 600 700 P(MW) ____P 0.2 0.4 0.6 0.8 1.0 1.2 P N

-20

-40

-60 Electronic Capacitorbank reactive-power regulation High-pass f ilter 2 -80

High- pass f ilter 1 -100

Reactive-Power Balance

UAC in p.u. (AC bus voltage) – cap. reactive-power reactive-power reactive-power reactive-power + ind converter AC filters reactors capacitors

2 2 2 QNetwork = + QConv – QFK*UAC +QL*UAC – QC*UAC

1010 Principle Arrangement of an HVDC Transmission Project 4

The Principle Arrangement of an HVDC Transmission Project is reflected on the Moyle Interconnector project. The HVDC stations between Northern Ireland and Scotland are operating with the following highlights: • Direct light triggered thyristor valves for the complete HVDC system, with 1872 thyristors in total, with 20% better reliability and all valve components free from oil. •Triple tuned AC filter in both stations. • Unmanned stations, fully automatic remote operation and automatic load schedule operation. • Hybrid optical ohmic shunt for DC current measuring unit. • Low noise station design for: – AC filter capacitor and reactors – converter transformer – converter valve water cooling system – DC hall with smoothing reactor •Station design for DC see/land cable with integrated return conductor and fibre optic cable for control and communication.

Date of contract 09/99 Delivery period 27 months

System Data Transmission capacity 2 x 250 MW System voltages 250 kV DC 275 kV AC Rated current 1000 A Transmission distance 63.5 km

Moyle Interconnector

Ballycronan Auchencrosh More

Overhead Line Undergroundround UndergroundU Cables CablesCa

Converter Station Converter Station

Existing 275-kV Undersea Cables Existing 275-kV Transmission Transmission System System Alternating Direct Alternating Current Current Current

Northern Ireland Scotland

1111 4.4 Principle Arrangement of an HVDC Transmission Project

DC HALL

LVE HALL VA

CONTROL BUILDING 250 kV DC

3 2 16 14 CONVERTER TRANSFORMER AREA

6 14 15 SHUNT CAPACITOR BANK5

10 8 9 250 DC Power Cable. HVDC Station Auchencrosh 63,5 km to HVDC Station Ballycronan More Smoothing Reactor Northern Ireland AC-Filter

Pole 1, 250 MW AC-Filter Thyristor Valves C-Shunt

Thyristor Valves AC-Filter

Pole 2, 250 MW AC-Filter Smoothing Reactor AC-Filter

AC Bus

12 11

ST AC FILTER

9 8 10

15 TT AC FILTER

275 kV AC SWITCHYARD

5 6 7 5 6 5 6

9 10 275 kV OHL 8 10 13 9 12

1 Quadruple Thyristor Valve 2 Converter Transformer 3 Air Core Smoothing Reactor 10 Current Transformer 4 Control Room and Control Cubicle 11 Voltage Transformer 5 AC Filter Capacitor 12 Combined Current-Voltage Transformer 6 AC Filter Reactor 13 Capacitive Voltage Transformer 7 AC Filter Resistor 14 Surge Arrester 8 Circuit Breaker 15 Earthing Switch 9 Disconnector 16 AC PLC Filter

13 5 Main Components

Components 5.1.1 Introduction Siemens is a leading supplier of HVDC The thyristor valves make the conversion systems all over the world. Our compo- from AC into DC and thus are the central nents are exceeding the usual quality component of any HVDC converter standards and are system-tailored to the station. The thyristor valves are of the needs of the grid. indoor type and air-insulated. Siemens has more than 30 years experience in the development and manufacturing of thyristor valves and has maintained the technical leadership by introducing new innovative concepts such as the corrosion- free water cooling and the self-protecting direct-light-triggered thyristor. This directly reflects in the high reliability of these valves.

valve valve valve

12-pulse group

Multiple valve unit – quadruple Valve valve branch

Fig. 5.1-1: Principle circuit diagram of a 12-pulse group consisting of three quadruple valves

Fig. 5.1-2: General arrangement of a 500 kV MVU (valve tower)

14 Thyristor Valves 5.1

5.1.2 Valve Design The modular concept of the Siemens thyristor valves permits different mechanical setups to best suit each application: single, double, quadruple valves or complete six-pulse bridges – either free – standing or suspended from the building structure. For seismic requirement reasons which exist in some regions of the world, the standard Siemens valves for long distance transmission are suspended from the ceiling of the valve hall. The suspension insulators are designed to carry the weight and additional loads originating for example from an unbal- anced weight distribution due to insulator failure, an earthquake or during maintenance. Connections between modules (piping of cooling circuit, fibre optic ducts, buswork, and suspension insulator fixtures) are flexible in order to allow stress-free deflections of the modules inside an MVU (multiple valve unit) structure. Figure 5.1-2 shows a typical quadruple valve tower for a 500 kV DC system. Each valve is made up of three modules. Four arresters, each related to one valve, are located on one side of the valve tower. Ease of access for maintenance purposes, if required, is another benefit of the Siemens valve design. By varying the number of thyristors per module and the number of modules per valve, the same design can be used for all transmission voltages that may be required.

15 5.1 Thyristor Valves

5.1.3 Thyristor Development 5.1.4 LTT (Light-Triggered Thyristor) Thyristors are used as switches and thus It has long been known that thyristors the valve becomes controllable. The can be turned on by injecting photons thyristors are made of highly pure mono- into the gate instead of electrons. The crystalline silicon. The high speed of use of this new technology reduces the innovation in power electronics technol- number of components in the thyristor ogy is directly reflected in the develop- valve up to 80%. This simplification ment of the thyristor. The high perform- results in increased reliability and ance thyristors installed in HVDC plants availability of the transmission system. today are characterized by silicon wafer With LTT technology, the gating light diameters of up to 5’’ (125 mm), blocking pulse is transmitted via a fibre optic cable through the thyristor housing directly to Fig. 5.1-4: Valve module with direct-light-triggered thyristor the thyristor wafer and thus no elaborate Thyristor Current Thyristor Blocking electronic circuits and auxiliary power (IdN) Voltage (UDRM) LTT 8 supplies are needed at high potential. kA kV The required gate power is just 40 mW. The forward overvoltage protection is integrated in the wafer. Further benefits 6 Thyristor Blocking 6 of the direct light triggering are the Voltage unlimited black start capability and the operation during system undervoltage or system faults without any limitations. 4 4 In case of electrically triggered thyristors (ETT), this is only possible if enough firing

Thyristor Current for energy is stored long enough on the Long-Distance Transmission thyristor electronics. 2 2 Direct light-triggered thyristors with Fig. 5.1-5: Silicon wafer and housing of a direct- integrated overvoltage protection (LTT) light-triggered thyristor, fibre optic cable for gating is a proven technology meanwhile and the Siemens standard. In 1997, it has 1970 1980 1990 2000 2003 been implemented successfully for the Light Pipe first time (Celilo Converter Station of the Fig. 5.1-3: Thyristor development Pacific Intertie). It shows excellent per- voltages up to 8 kV and current carrying formance and no thyristor failures or capacities up to 4 kA DC. Thus no parallel malfunction of the gating system have Cu Si thyristors need to be installed in today’s been recorded. BPA has emphasized its confidence in this technology in 2001 by HVDC systems for handling the DC Cu Mo current. The required DC system voltages awarding Siemens the contract to replace are achieved by a series connection of all mercury arc valves with direct-light- a sufficient number of thyristors. triggered thyristor valves. Furthermore, this valve technology is used for the Fig. 5.1-6: The optical gate pulse is transmitted Moyle Interconnector (2 x 250 MW), directly to the thyristor wafer which went into service in 2001 and is on contract for the 3000-MW, ± 500-kV Guizhou-Guangdong system. Monitoring of the thyristor performance is achieved by a simple voltage divider circuit made from standard off-the-shelf resistors and capacitors; monitoring signals are transmitted to ground potential through a dedicated set of fibre optic cables as for the ETT. However, all electronic circuits needed for the evaluation of performance are now located at Cu = Copper ground potential in a protected environ- Si = Silicon ment, further simplifying the system. Mo = Molybdenum The extent of monitoring is the same as LTT=Light-triggered thyristors for the ETT. ETT=electrically triggered It can be expected that this technology thyristors will become the industry standard in HVDC thyristor valves of the 21st century, paving the way towards maintenance- 16 free thyristor valves. 5.1.5 Valve Cooling 5.1.6 Flame Resistance Siemens has used the parallel water Much effort has been invested by cooling principle for more than 25 years. Siemens to minimize the fire risk: No corrosion problems have ever been encountered. • All oil has been eliminated from the valve and its components. Snubber The thyristors are stacked in the module capacitors and grading capacitors use with a heat sink on either side. The water SF6 as a replacement for impregnating connection to the heat sinks can be oil. designed in parallel or series as shown • Only flame-retardant and self-extin- in figure 5.1-7. The parallel cooling circuit guishing plastic materials are used. provides all thyristors with the same •A wide separation between the cooling water temperature. This allows a modular units ensures that any local better utilization of the thyristor capability. overheating will not affect neighbouring Siemens makes use of this principle, components. which offers the additional advantage • Careful design of the electrical that electrolytic currents through the heat connections avoids loose contacts. sinks – the cause for electrolytic corrosion – can be avoided by placing grading The past has shown that Siemens HVDC electrodes at strategic locations in the installations have never been exposed water circuit. Siemens water cooling also to a hazardous valve fire. The tests does not require any de-oxygenizing performed on actual components and equipment. samples in the actual configuration as used in the valve indicate that the improved design indeed is flame-retardant and the risk of a major fire following a fault is extremely low or even non- existent.

Fig. 5.1-8: Converter valves d Sylmar HVDC station, Los Angeles, USA b a c

a c b

Fig. 5.1-7: Piping of module cooling circuit – parallel flow (top); series flow (bottom) a) thyristor; b) heat sink; c) connection piping; d) manifold

1517 5.2 Converter Transformer

Siemens supplies transformers which Project: Tian Guang meet all requirements concerning power, HVDC bipolar long- voltage, mode of operation, low noise distance transmission level, connection techniques, type of PN = 2 x 900 MW cooling, transport and installation. They Ud = ± 500 kV also comply with special national design Transformers: requirements. SN = 354/177/177 MVA 1-ph/3-w unit All over the world, power transformers UAC = 220 kV from Nuremberg enjoy a great reputation. What the Nuremberg plant manufactures reflects today’s state of the art and testifies to the highest levels of quality and reliability. Our quality management system is certified to DIN 9001, the world’s most stringent standard. Our accredited test laboratories likewise meet the latest specifications.

Fig. 5.2-1: Converter transformer for the Tian Guang HVDC project during type test

Converter transformer for the Three Gorges HVDC project 284 MVA, 1-ph/2-w unit

1618 5.2.1 Functions of the HVDC 5.2.2 Tr ansformer Design Variations Converter Transformer There are several aspects which play a The converter transformers transform role in selecting the transformer design: the voltage of the AC busbar to the required entry voltage of the converter. Transportation Weight and Dimensions The 12-pulse converter requires two In systems of high power, weight can 3-phase systems which are spaced apart be an important consideration, in from each other by 30 or 150 electrical particular where transportation is difficult. degrees. This is achieved by installing a The relative transportation weights of transformer on each network side in the the 4 major design types are vector groups Yy0 and Yd5. approximately as follows: At the same time, they ensure the voltage insulation necessary in order to Single-phase – two-winding transformer 1 make it possible to connect converter Single-phase – three-winding transformer 1.6 bridges in series on the DC side, as is necessary for HVDC technology. The Three-phase – two-winding transformer 2.2 transformer main insulation, therefore, Three-phase – three-winding transformer 3.6 is stressed by both the AC voltage and the direct voltage potential between The transport dimension and the weight valve-side winding and ground. The of the converter transformer depends converter transformers are equipped with on the limitations for street, railway and on-load tap-changers in order to provide shipping, especially in the case of the correct valve voltage. bridges, subways and tunnels.

Transformer Rating 5.2.3 HVDC Makes Special Demands on Transformers STrafo Rec (6-pulse) = √2 * IdN * Usec Rec HVDC transformers are subject to I nominal DC current dN operating conditions that set them apart U Transformer-voltage sec Rec from conventional system or power valve side (Rectifier) transformers. These conditions include: S = 2 * I * U Trafo Inv (6-pulse) √ dN sec Inv • Combined voltage stresses Usec Inv Transformer voltage • High harmonics content of the valve side (inverter) operating current • DC premagnetization of the core The valve windings which are connected to the rectifier and the converter circuit are subject to the combined load stress of DC and AC voltage. Added to this stress are transient voltages from outside caused by lightning strikes or switching operations. The high harmonics content of the operating current results from the virtually quadratic current blocks of the power converter. The odd-numbered harmonics with the ordinal numbers of 5, 7, 11, 13, 17 … cause additional losses in the windings and other structural parts.

1719 5.2 Converter Transformer

5.2.4 Main Components of the Windings Bushings Converter Transformer The large number of parameters con- Compared to porcelain, composite Core cerning transport limitations, rated power, bushings provide better protection transformer ratio, short-circuit voltage, against dust and debris. A 15% higher HVDC transformers are normally single- and guaranteed losses require significant DC voltage testing level compared to the phase transformers, whereby the valve flexibility in the design of windings. windings underscores the particular windings for the star and delta connection safety aspect of these components. are configured either for one core with In concentric winding arrangements, star at least two main limbs or separately for or delta valve windings lying directly on Special Tests for HVDC Transformers two cores with at least one main limb, the core have proven optimal in many depending on the rated power and the cases. The line winding, normally with a Special tests for verifying operating system voltage. Appropriately sized return tapped winding, is then mounted radial functionality are required for HVDC limbs ensure good decoupling for a outside this core configuration. transformers. The applicable international combined arrangement of windings. standards are subject to constant further The valve windings with high insulation development. Separate tests with DC The quality of the core sheets, the levels and a large portion of current voltage, switching and lightning impulse lamination of the sheets, and the nominal harmonics make particular demands on voltages cover the range of different induction must all conform to special the design and the quality of the winding voltage loads. The 2-MV DC voltage requirements covering losses, noise level, manufacturing. Together with its generator in the Nuremberg Transformer over-excitation, etc. Special attention pressboard barriers, each limb set, in- Plant is well-suited for all required DC must be paid to the DC premagnetization cluding a valve, an overvoltage and a voltage and reverse poling tests. The of the core due to small asymmetries tapped winding, forms a compact unit, most important criterion is partial during operation and stray DC currents which is able to cope with the demand discharge. A maximum of 10 discharges from the AC voltage network. The effects made by voltage stress, loss dissipation, over 2000 pC during the last 10 minutes of DC premagnetization must be and short-circuit withstand capability. of the test is permitted. compensated by appropriate design and manufacturing efforts (e.g. additional core Tank cooling ducts, avoidance of flux pinching in the core sheet). The unconventional tank design in HVDC transformers result from the following requirements: •The valve-side bushing should extend into the valve hall •The cooling system is mounted on the opposite side to facilitate rapid transformer exchange For HVDC transformers with delta and star valve winding in one tank, the valve bushing must be arranged so that their ends conform to the geometry of the thyristor valve towers. This frequently leads to very high connection heights and the need to mount the oil expansion tank at a significant height. In close cooperation with the equipment design department, the engineering specialists at the Nuremberg Transformer Plant have always been able to find a design suited to every customer requirement.

20 Smoothing Reactor 5.3

5.3.1 Functions of the Smoothing smoothing reactors is often selected The wall bushing in composite design Reactor in the range of 100 to 300 mH for long- is the state-of-the-art technology which distance DC links and 30 to 80 mH for provides superior insulation perform- •Prevention of intermittent current back-to-back stations. ance. • Limitation of the DC fault currents •Prevention of resonance in the 5.3.3 Arrangement of the DC circuit Smoothing Reactor •Reducing harmonic currents including limitation of telephone interference In an HVDC long-distance transmission system, it seems quite logical that the Prevention of intermittent current smoothing reactor will be connected in series with the DC line of the station pole. The intermittent current due to the This is the normal arrangement. current ripple can cause high over-volt- However in back-to-back schemes, the ages in the transformer and the smoothing smoothing reactor can also be connected reactor. The smoothing reactor is used to the low-voltage terminal. to prevent the current interruption at minimum load. 5.3.4 Reactor Design Alternatives Limitation of the DC fault current There are basically two types of reactor design: The smoothing reactor can reduce the • Air-insulated dry-type reactors fault current and its rate of rise for • Oil-insulated reactors in a tank commutation failures and DC line faults. The reactor type should be selected taking This is of primary importance if a long the following aspects into consideration: DC cable is used for the transmission. • Inductance For an overhead line transmission, the • Costs current stress in valves is lower than the • Maintenance and location of spare stress which will occur during valve short units circuit. • Seismic requirements Fig. 5.3-1: Oil-insulated smoothing reactor – Prevention of resonance in the An advantage of the dry-type reactor is Three Gorges project DC circuit that maintaining spare units (to the extent • Inductance: 270 mH necessary) is not very expensive because • Rated voltage: 500 kV DC The smoothing reactor is selected to avoid they usually consist of several partial • Rated current: 3000 A DC resonance in the DC circuit at low order coils. However for very large inductances harmonic frequencies like 100 or 150 Hz. it is possible to have more than one unit This is important to avoid the ampli- and it could be a problem if much space fication effect for harmonics originally is not available. from the AC system, like negative sequence and transformer saturation. In high seismic regions, setting them on post-insulators or on an insulating plat- Reducing harmonic currents form is a possible problem. Oil-insulated including limitation of telephone smoothing reactors are then the preferred interference solution. Limitation of interference coming from The oil-insulated reactor is economical 2 the DC overhead line is an essential for very high power (Id * Ldr). function of the DC filter circuits. However, It is the best option for regions with high the smoothing reactor also plays an seismic requirements. important role to reduce harmonic currents acting as a series impedance. One bushing of the oil-insulated smoothing reactor penetrates usually 5.3.2 Sizing of the smoothing Reactor into the valve hall, while the other bushing is normally in a vertical position. While the current and voltage rating of For the air-insulated dry-type smoothing the smoothing reactor can be specified reactor, a wall bushing is needed to based on the data of the DC circuit, the connect with the valves. inductance is the determining factor in sizing the reactor.Taking all design Fig. 5.3-2: Air-insulated smoothing aspects above into account, the size of reactor – Tian Guang project • Inductance: 150 mH • Rated voltage: 500 kV DC • Rated current: 1800 A DC

21 5.4 Harmonic Filters

The filter arrangements on the AC side Fig. 5.4.1-1: of an HVDC converter station have two Harmonic Order Different harmonic main duties: filter types • to absorb harmonic currents generated 1000 by the HVDC converter and thus to 800 reduce the impact of the harmonics 600 on the connected AC systems, like AC 400 ilter Impedance ilter

voltage distortion and telephone F 200 interference 0 • to supply reactive power for compen- Single- 01020304050 sating the demand of the converter tuned station Each filter branch can have one to three Harmonic Order tuning frequencies. Figure 5.4.1-1 shows different harmonic filter types with their impedance frequency characteristics. 1000 800 600 400 200 ilter Impedance ilter

F 0 Double- 01020304050 tuned

Harmonic Order

800 600 400 200 ilter Impedance ilter

F 0 Triple- 01020304050 tuned

Fig. 5.4.1-2: AC filters and capacitor banks of Gezhouba/Shanghai

22 AC Harmonic Filter 5.4.1

5.4.1.1 Design Criteria for AC Filters related to the harmonic voltage on the There are basically two methods to converter station busbar. The purpose of include the network impedance in the Reactive Power Requirements the filter circuit is to provide sufficiently filter calculations: low impedances for the relevant harmo- The reactive power consumption of an • to calculate impedance vectors for all HVDC converter depends on the active nic components in order to reduce the harmonic voltages to an acceptable level. relevant harmonics and grid conditions power, the transformer reactance and • to assume locus area for the imped- the control angle. It increases with in- The acceptance criteria for the harmonic ance vectors creasing active power. A common re- distortion depend on local conditions and quirement to a converter station is full regulations. A commonly used criterion The modelling of a complete AC network compensation or overcompensation at for all harmonic components up to the with all its components is very complex rated load. In addition, a reactive band for 49th order is as follows: and time-consuming. For this reason, the the load and voltage range and the locus method is very often used. It is permitted voltage step during bank Dn individual harmonic voltage based on a limited number of measure- switching must be determined. These distortion of order n in percent of ments or calculations. Different locus factors will determine the size and the fundamental AC busbar voltage areas for different harmonics or bands number of filter and shunt capacitor (typical limit 1%) are often determined to give a more banks. precise base for the harmonic perform- Drms total geometric sum of individual ance calculation. Harmonic Performance Requirements voltage distortion Dn (typical limit 2%) A typical locus area is shown in HVDC converter stations generate fig. 5.4.1-4. It is assumed that the im- characteristic and non-characteristic The BTS Telephone Interference Factor pedance vector will be somewhere inside harmonic currents. For a twelve-pulse (TIF) and the CCITT Telephone Harmonic the perimeter of the coloured area. converter, the characteristic harmonics Form Factor (THFF) are determined with weighted factors in order to evaluate the The impedance vector of the filter is are of the order n = (12 * k) ± 1 (k = 1,2,3 ...). These are the harmonic components voltage distortion level on the AC busbar transformed into the Y plane for each that are generated even during ideal with respect to the expected interference harmonic frequency. conditions, i.e. ideal smoothing of the level in nearby analogue telephone systems. The IT product is a criterion for With both the network and the filter direct current, symmetrical AC voltages, impedances plotted in the admittance transformer impedance and firing angles. harmonic current injected into AC overhead lines. The criteria based on tele- plane, the shortest vector between the The characteristic harmonic components filter admittance point and the network are the ones with the highest current phone interference are in many cases irrelevant, because modern digital tele- admittance boundary gives the lowest level, but other components may also possible admittance value for the parallel be of importance. The third harmonic, phone systems are insensitive to harmonic interference. combination of the network and the filter. which is mainly caused by the negative This value is used to determine the sequence component of the AC system, Network Impedance highest possible harmonic voltage. will in many cases require filtering. An equivalent circuit for determination The distortion level on the AC busbar of harmonic performance is given in depends on the grid impedance as well figure 5.4.1-3. The most commonly used as the filter impedance. An open circuit criteria for harmonic performance are model of the grid for all harmonics is not on the safe side. Parallel resonance between the filter impedance and the grid impedance may create unacceptable amplification of harmonic components for which the filters are not tuned. For Ymin = 1/Rmax Ymax = 1/Rmin Ih ZF Uh ZN this reason, an adequate impedance model of the grid for all relevant harmonics is required in order to optimize the filter design. Yf Harmonic Filter Network Current Impedance Impedance Source Yres

Fig. 5.4.1-4 Circle of network admittance and Fig. 5.4.1-3: Equivalent circuit for calculation the resonance conditions of harmonic voltages and currents in the AC system

Ih = harmonic source current Zf = filter impedance ZN = network impedance Uh = harmonic voltage

23 5.4.1 AC Harmonic Filter

The selective resonance method repre- Requirements to Ratings • Filter Energization sents a reasonable compromise. It takes The AC filter is assumed to be Steady-State Calculation into consideration the fact that the energized at the moment for the highest voltage distortion (highest har- The voltage and current stresses of AC maximum AC bus peak voltage. This monic voltage) occurs with a parallel filters consist of the fundamental case is decisive for the inrush currents resonance between filter and AC network. frequency and harmonic components. of AC filters. It is unrealistic however, to assume that Their magnitudes depend on the AC Fault Recovery after Three-Phase such a parallel resonance takes place at • system voltage, harmonic currents, Ground Fault all frequencies. Normally it is sufficient operating conditions and AC system Various fault-clearing parameters to consider in the calculation of total impedances. The rating calculations are distortion and TIF value only two maxi- should be investigated to determine carried out in the whole range of the maximum energy stresses for AC mum individual distortions from the operation to determine the highest resonance calculation. The AC network is filter arresters and resistors. The worst- steady-state current and voltage stresses case stresses are achieved if the assumed to be open for the remaining for each individual filter component. harmonic currents. HVDC converters are blocked after fault initiation, while the AC filters Tr ansient Calculation The filter calculations must reflect de- remain connected to the AC bus after tuning caused by AC network frequency The objective of the transient rating fault clearing and recovery of the AC deviations and component parameter calculation is to determine the highest system voltage. In this case, a tempo- deviations. Production tolerances, tempe- transient stresses for each component rary overvoltage with high contents of rature drift and failure of capacitor ele- of the designed filter arrangement. The non-characteristic harmonics will occur ments are the main contributors to results of the transient calculation should at the AC bus due to the effects of parameter deviations. contain the voltage and current stresses load rejection, transformer saturation for each component, energy duty for filter and resonance between filter and resistors and arresters, and the insulation AC network at low frequency. levels for each filter component. To calculate the highest stresses of both lightning and switching surge type, different circuit configurations and fault cases should be studied: • Single-Phase Ground Fault The fault is applied on the converter AC bus next to the AC filter. It is assumed that the filter capacitor is charged to a voltage level corresponding to the switching impulse protective level of the AC bus arrester. •Switching Surge For the calculation of switching surge stresses, a standard wave of 250/2500 µs with a crest value equal to the switching impulse protective level of the AC bus arrester is applied at the AC converter bus.

Fig. 5.4.2-1 DC filter of Guangzhou/China 24 DC Harmonic Filter 5.4.2

5.4.2.1 DC Filter Circuits 5.4.2.2 Design Criteria for DC Filter The equivalent disturbing current com- Circuits bines all harmonic currents with the aid Harmonic voltages which occur on the of weighting factors to a single inter- DC side of a converter station cause AC The interference voltage induced on the ference current. With respect to tele- currents which are superimposed on the telephone line can be characterized by phone interference, it is the equivalent direct current in the transmission line. the following equation: to the sum of all harmonic currents. It These alternating currents of higher also encompasses the factors which frequencies can create interference in m 2 determine the coupling between the neighbouring telephone systems despite Ieq = ∑(Hµ * Cµ * Iµ(x)) HVDC and telephone lines: limitation by smoothing reactors. √ 1 DC filter circuits, which are connected V = Z I • Operating mode of the HVDC system in parallel to the station poles, are an in(x) * eq (bipolar or monopolar with metallic or effective tool for combating these pro- where ground return) blems. The configuration of the DC filters • Specific ground resistance at point x very strongly resembles the filters on the AC side of the HVDC station. There are Vin(x) = Interference voltage on the The intensity of interference currents is several types of filter design. Single and telephone line at point x strongly dependent on the operating multiple-tuned filters with or without the (in mV/km) condition of the HVDC. In monopolar high-pass feature are common. One or Hµ =Weighting factors which reflect operation, telephone interference is several types of DC filter can be utilized the frequency dependence of significantly stronger than in bipolar in a converter station. the coupling between tele- operation. phone and HVDC lines Cµ = “C message“ – weighting factors Iµ(x) =Resulting harmonic current of the ordinal number µ in the HVDC line at point x as the vector sum of the currents caused by the two HVDC stations Ieq = Psophometric weighted equivalent disturbing current Z=Mutual coupling impedance between the telephone and HVDC lines

25 5.4.3 Active Harmonic Filter

Active filters can be a supplement to A transformer matches the voltage and passive filters due to their superior per- current levels at the converter output formance. They can be installed on the and provides the required insulation level. DC side or on the AC side of the convert- The goal of the scheme is to inject har- er. The connection to the high-voltage monics in the network with the same system is achieved by means of a passive amplitude and the opposite phase of the filter, forming a so-called hybrid filter. This harmonics at the measurement point in arrangement limits the voltage level and order to cancel them. the transient stresses on the active part, so that comparatively low equipment The filter for AC application comprises ratings can be used. Appropriate design three single-phase systems controlled allows the exploitation of the positive by a common digital control system. A characteristics of both passive and active major difference is the measurement: filters. Additionally, the passive part instead of measuring the line current, can be used as a conventional passive the active filter at Tjele measures and filter if the active part is by-passed for eliminates harmonics at the 400 kV AC Fig. 5.4.3-1: Active DC filter on site busbars of the station. This has the ad- (Tian Guang HVDC project) maintenance purposes. vantage that the harmonic control requires just one measurement point, compared to a current measuring scheme, Main Components which would require to measure the 400 kV Bus No.Component current at several points and combining the measured signals. The other advan-

Capacitive Existing 1 IGBT converter tage is that the active filter works just Potential Optical Passive Interface 2Reactor for inductivity like a passive filter ideally should do, i.e. Transformer Filter From the ZF2 adapting eliminating the voltage in the bus, thus Other Phases representing no change in philosophy. Underground 3 Thyristor switch for con- Active AC Filter Cable To/From Main verter overvoltage and The active filter is fully assembled in a Common Simadyn D Components overcurrent protection transportable container and is tested at cubicle of the Other Control and Auxiliary (control and the factory as a complete system before protection) Phases (in 4Transformer Equipment the Container) shipping. Fig. 5.4.3-5 shows the installed 2 5 Low-pass filter active AC filter (in the container) at the 1 34 5 67 6Vacuum switch Tjele substation. IGBT LP conv. 8 9 7 ZnO arrester 8 Isolators and grounding switches 9 LC branch for deviating Main Components (one Set for Each Phase) the 50-Hz current component Fig. 5.4.3-2: Single-line diagram of the active AC filter. All phases have the same topology. The Siemens active filters use voltage- sourced IGBT converters with a high switching frequency to produce an output voltage up to approximately 700 Vpeak, containing harmonics up to the 50th as required. A powerful high-speed control and protection system processes the currents and/or voltages measured at the network by appropriate sensors and produces the control pulses for the IGBT’s.

26 One harmonic controller is dedicated to each harmonic selected for elimination by the action of the active filter. In these harmonic controllers, the particular Voltage on the To the IGBT harmonic is isolated and expressed by 400-kV Busbar Converter a complex signal in the frequency domain. This is done through multiplication by sin (hωt) and cos (hωt), where h is the Optical Self-Tuning IGBT order of the harmonic, ω the network Input system control angular frequency and t the time. These two orthogonal signals are produced by LP PI a module synchronized by the funda- Filter Controller Σ mental component of the filter current. The signal pair obtained after the mentioned multiplication and filtering To Other From Other feeds a complex controller with PI Harmonic Harmonic characteristic. The output of the controller Controllers Synchronization cos (hωt) Controllers sin (hωt) is then shifted back to the time domain by multiplication by cos (hωt) and sin (hωt). Harmonic Controller The process is essentially linear, so that all harmonic controllers can operate simultaneously and the sum of all har- Fig. 5.4.3-3: Principle block diagram of the harmonic control monic controller outputs gives the wave- form required by the active filter. This signal is then given to the IGBT control module, which includes a pulse width modulator besides functions for protection and supervision of the converter.

Harmonic voltages at the 400-kV bus …with active filter control (23th, 25th, (L1) without and… 35th, 47th and 49th harmonics) Fig. 5.4.3-4: Plots from measurement: left without, right with active filter control

Fig. 5.4.3-5: Installation of the active AC filter, 400-kV substation Tjele (Denmark)

27 5.5 Surge Arrester

Siemens surge arresters are designed optimally to the following requirements:

Excellent pollution performance for Arrester Voltage Referred Rated Voltage ÛR coastal and desert regions or in areas to Continuous Operating Continuous Operating Voltage Û/ÛC with extreme industrial air pollution. Voltage ÛC High mechanical stability, e.g. for use in seismic zones. 2 Extremely reliable pressure relief behaviour for use in areas requiring special protection.

What is more, all Siemens surge arresters are sized for decades and the material 1 used provides a contribution towards the 20 °C protection of the environment. 115 °C The main task of an arrester is to protect 150 °C the equipment from the effects of overvoltages. During normal operation, it should 0 have no negative effect on the power 10-4 10-3 10-2 10-1 1 10 102 103 104 system. Moreover, the arrester must be Current through Arrester I [A] able to withstand typical surges without a incurring any damage. Non-linear resistors with the following properties fulfil these Fig. 5.5-1: Current/voltage characteristics of a non-linear MO arrester requirements: • Low resistance during surges so that overvoltages are limited • High resistance during normal operation in order to avoid negative effects on the power system and • Sufficient energy absorption capability for stable operation MO (Metal Oxide) arresters are used in medium-, high-and extra-high-voltage power systems. Here, the very low protection level and the high energy absorption capability provided during switching surges are especially important. For high voltage levels, the simple construction of MO arresters is always an advantage.

Arresters with Polymer Housings Fig. 5.5-2 shows two Siemens MO arresters with different types of housing. In addition to what has been usual up to now – the porcelain housing – Siemens offers also the latest generation of high- voltage surge arresters with polymer housing.

Fig. 5.5-2: Measurement of residual voltage on porcelain-housed (foreground) and polymer-housed (background) arresters

28 Fig. 5.5-3 shows the sectional view of For terminal voltage lower than the such an arrester. The housing consists of permissible maximum operating voltage a fibre-glass-reinforced plastic tube with (MCOV), the arrester is capacitive and insulating sheds made of silicon rubber. carries only few milli-amps. Due to its The advantages of this design which has extreme non-linear characteristics, the the same pressure relief device as an arrester behaves at higher voltages as arrester with porcelain housing are low-ohmic resistor and is able to dis- absolutely safe and reliable pressure charge high current surges. Through relief characteristic, high mechanical parallel combination of two or more strength even after pressure relief and matched arrester columns, higher energy excellent pollution-resistant properties. absorption capability of the ZnO arrester The very good mechanical features mean can be achieved. that Siemens arresters with polymer housing (type 3EQ/R) can serve as post Routine and type tests have been insulators as well. The pollution-resistant determined in accordance with the properties are the result of the water- international standards: repellent effect (hydrophobicity of the IEC 60060 High-voltage test techniques silicon rubber). IEC 60071 Insulation coordination The polymer-housed high-voltage arrester IEC 60099 Surge arresters design chosen by Siemens and the high- quality materials used by Siemens provide a whole series of advantages including long life and suitability for outdoor use, high mechanical stability Fig. 5.5-3: Cross-section of a polymer-housed and ease of disposal. arrester

Flange with Gas Diverter Nozzle Seal

Pressure Relief Diaphragm Compressing Spring

Metal Oxide Resistors

Composite Polymer Housing FRP Tube/Silicon Sheds

29 5.5 Surge Arrester

AC Filter Bank 9 10 DC to DC line 10 1 DC Aa AC V D D 2 Filter 4 AC Bus 5 11

A V DC 12 3 7 C Filter V

Fdc2 6 Fac1 Fac2 Fdc1 A V 8 Neutral

Valve Hall Boundary E Cn E Neutral 8

Arrester Type Location Main Task

AC bus arrester ’A’ The ZnO arrester will be installed close to Limit the overvoltages on the primary and secondary the converter transformer line side bushing side of the converter transformer

AC filter bus The ZnO arrester will be installed at the Protect the AC filters busbar against arrester ‘Aa’ busbar of the AC filter banks lightning surges

Valve-arrester ‘V’ 3-pulse commutation group The main events to be considered with respect to arrester discharge currents and energies are: a) Switching surges from the AC system through converter transformer b) Ground fault between valve and HV bushing of converter transformer during rectifier operation

Converter group 12-pulse converter group Protection against overvoltages from the arrester ‘C’ AC and DC side

DC bus arrester ‘D’ At the HV smoothing They will protect the smoothing reactor and the reactor and at the DC lines converter station (e.g. DC switchyard) against overvoltages coming from the DC side

Neutral DC bus Neutral DC bus The neutral bus arresters protect the LV terminal of arrester ‘E’ the12-pulse group and the neutral bus equipment

AC filter arrester AC filter The operating voltage for the AC filter arresters consists ‘Fac‘ of low fundamental frequency and harmonic voltages. Overvoltages can occur transiently during faults

DC filter arrester DC filter The operating voltage for the DC filter arresters consists ‘Fdc‘ of low DC component and harmonic voltages. Overstresses may occur transiently during DC bus fault to ground

30 DC Transmission Circuit 5.6

5.6.1 DC Transmission Line 5.6.1.1 To wers DC transmission lines could be part of Such DC transmission lines are mechani- overall HVDC transmission contract either cally designed as it is practice for normal within a turnkey package or as separately AC transmission lines; the main contracted stand-alone item, later inte- differences are: grated into an HVDC link. •The conductor configuration •The electric field requirements As an example of such a transmission •The insulation design line design, an existing bipolar tower for the 300-kV link between Thailand and 5.6.1.2 Insulation Malaysia is shown in Fig.5.6.1-1 and a bipolar AC transmission tower of Tian The most critical aspect is the insulation Guang is shown in Fig. 5.6.1-2. design and therefore this topic is described more detailed below: For DC transmission lines, the correct insulation design is the most essential subject for an undisturbed operation during the lifetime of the DC plant.

Design Basics •The general layout of insulation is based on the recommendations of IEC 60815 which provides 4 pollution classes. •This IEC is a standard for AC lines. It has to be observed that the creepage distances recommended are based on the phase-to-phase voltage (UL-L). When transferring these creepage distances recommended by IEC 60815 to a DC line, it has to be observed that the DC voltage is a peak voltage pole to ground value (UL-G). Therefore, these creepage distances have to be multi- plied by the factor √3. • Insulators under DC voltage operation Fig. 5.6.1-1: are subjected to more unfavourable DC transmission line conditions than under AC due to higher (bipolar tower 300-kV link) collection of surface contamination caused by constant unidirectional electric field. Therefore, a DC pollution factor as per recommendation of CIGRE (CIGRE-Report WG04 of Cigre SC33, Mexico City 1989) has to be applied. The correction factors are valid for porcelain insulators only. When taking composite insulators into consideration, additional reduction factors based on the FGH report 291 “Oberflächenverhalten von Freiluftgeräten mit Kunststoffge- häusen“ must be applied.

Fig. 5.6.1-2: DC transmission tower (bipolar) Tian Guang (South China)

31 5.6.1 DC Transmission Line

Types of Insulators Long-Rod Porcelain Type Composite Long-Rod Type There are 3 different types of insulators Positive Aspects: Positive Aspects: applicable for DC transmission lines: •Long-term experience/track record • Small number of insulators in one string • Cap and pin type • Good mechanical strength • Up to 400 kV per unit possible •Long-rod porcelain type •Puncture-proof • Good mechanical strength, no chipping • Composite long-rod type • Good self-cleaning ability of sheds possible •Less intermediate metal parts •Very light – easy handling during In detail: • Due to caps on both insulator ends not construction and maintenance, logistical Cap and Pin Type subjected to pin corrosion because of advantages in areas with poor access low track current density •Puncture-proof Positive Aspects: • Moderate price • Good self-cleaning behaviour – •Long-term experience/track record hydrophobicity of surface which offers • Good mechanical strength Negative Aspects: advantages of less creepage distance •Vandalism-proof • Heavy strings up to pollution class II • Flexibility within the insulator string • String not very flexible •Very good RIV and corona behaviour Negative Aspects: • Under extreme vandalism failure of • Good resistance against vandalism •Very heavy strings string possible • Shorter insulator string length • Insulator not puncture-proof •Very competitive price •Poor self-cleaning ability •Loss of strength/reliability due to Negative Aspects: corrosion of pin in polluted areas caused •Relatively short track record in DC appli- by high track current density (this is cation (since 1985 first major application extremely important for DC lines) in the USA) • Many intermediate metal parts •Less tracking resistance against flash- • High RIV and corona level over (can be improved by means of •For DC applications, special shed design corona rings) and porcelain material necessary •Very expensive

Example/Comparison of Insulator Application for a 400 kV Transmission Line Cap and Pin Porcelain Long-Rod Composite Long-Rod

Insulator string length 5270 mm 5418 mm 4450 mm 31 insulators 4 insulators 1 insulator

Creepage per unit 570 mm 4402 mm 17640 mm

Weight of string 332 kg 200 kg 28 kg

Breaking load 160 kN 160 kN 160 kN

32 DC Cable 5.6.2

5.6.2.1 General Application for 2) Oil-Filled Cable 2) Lapped Thin Film Insulation DC Cables In comparison to mass-impregnated cables, the conductor is insulated by As insulating material a lapped non- An important application for HVDC are paper impregnated with a low-viscosity impregnated thin PP film is used instead transmission systems crossing the sea. oil and incorporates a longitudinal duct of the impregnated materials. The tests Here, HVDC is the preferred technology to permit oil flow along the cable. Oil- for the cable itself are completed. Now to overcome distances > 70 km and trans- filled cables are suitable for both AC and the tests for the accessories such as mission capacities from several hundred DC voltages with DC voltages up to joints are under process. to more than a thousand MW (for bipolar 600 kV DC and great sea depths. Due This type of cable can sustain up to 60% systems). For the submarine transmission to the required oil flow along the cable, higher electrical stresses in operation, part, a special cable suitable for DC the transmission line lengths are however making it suitable for very long and deep current and voltage is required. limited to <100 km and the risk of oil submarine cables. leakage into the environment is always 5.6.2.2 Different Cable Types Another area of development are the subject to discussions. cable arrangements. For monopolar For HVDC submarine cables there are transmission systems, either the return different types available. 5.6.2.3 Future Developments for path was the ground (’ground return‘) or HVDC Cables a second cable. The first solution always 1) Mass-Impregnated Cable provokes environmental concerns where- This cable type is used in most of Most of the research and development activities for new cable types are done as the second one has excessive impact the HVDC applications. It consists on the costs for the overall transmission of different layers as shown in with the insulation material. These include: scheme. Fig. 5.6.2.2-1. Therefore, a new cable was developed The conductor is built of stranding copper 1) XLPE with an integrated return conductor. The cable core is the traditional design for a layers of segments around a central To overcome the disadvantages of the circular rod. The conductor is covered by mass-impregnated cable and the return above mentioned cable types, extensive conductor is wound outside the lead oil and resin-impregnated papers. The R&D was conducted by the cable inner layers are of carbon-loaded papers sheath. The conductor forms also part of suppliers. The result is the XLPE cable. the balanced armour, together with the whereas the outer layer consists of XLPE means ‘cross-linked polyethylene‘ copper-woven fabrics. flat steel wire layer on the outside of the and forms the insulation material. The return conductor insulation. The fully impregnated cable is then lead- conductor is the segmented copper sheathed to keep the outside environ- This cable type was installed between conductor insulated by extruded XLPE Scotland and Northern Ireland for ment away from the insulation. The layers. The insulation material is suitable next layer is the anti-corrosion protection 250 kV and 250 MW. R&D is ongoing to for a conductor temperature of 90°C and which consists of extruded polyethylene. increase the voltage as well as the capa- a short-circuit temperature of 250°C. Around the polyethylene layer galvanized city of the cable with integrated return Although the main application for XLPE conductor. steel tapes are applied to prevent the cables is the land installation and the off- cable from permanent deformation shore industry, XLPE with extruded during cable loading. Over the steel tapes insulation material for HVDC systems of a polypropylene string is applied followed lower transmission capacities are under by galvanized steel wire armour. development. The technology is available for voltages up to 500 kV and a transmission capacity of up to 800 MW in one cable with installation depths of up to 1000 m under sea level and nearly unlimited transmission lengths. The capacity of mass-impregnated cables is limited by the conductor temperature which results in low overload capabilities.

1Conductor of copper-shaped wires 2 Insulation material 3 Core screen 4Lead alloy sheath 5Polyethylene jacket 6Reinforcement of steel tapes 7Bedding 8 Armour of steel flat wires

Fig. 5.6.2.2-1: Mass-impregnated cable

33 5.6.3 High Speed DC Switches

5.6.3.1 General 5.6.3.2 Types and Duties of the High-Speed DC Switches Like in AC substations, switching devices Type Duties are also needed in the DC yard of HVDC stations. One group of such devices can HSNBS The HSNBS must commutate some be characterized as switches with direct (High-Speed Neutral Bus Switch) direct current into the ground electrode current commutation capabilities, path in case of faults to ground at the commonly called ”high-speed DC station neutral. switches”. HSGS The HSGS is needed to connect the (High-Speed Ground Switch) station neutral to the station ground grid Siemens standard SF6 AC circuit-breakers of proven design are able to meet the if the ground electrode path becomes requirements of high-speed DC switches. isolated. MRTB If one pole of a bipolar system has to be (Metallic Return Transfer Breaker) blocked, monopolar operation of the second pole is achieved automatically, HVDC Overhead Line but with return current through ground (refer to Fig. 5.6.3-1). If the duration of Electrodes ground return operation is restricted, an alternate mode of monopolar operation System 2 System System 1 System is possible if the line of the blocked pole AC AC HVDC can be used for current return. This mode Overhead Line is called metallic return (refer to Fig. 5.6.3-1). The MRTB is required for the transfer from ground to metallic return Bipolar without interruption of power flow. GRTS The GRTS is needed for the retransfer (Ground Return Transfer Switch) from metallic return to bipolar operation HVDC via ground return, also without Overhead Line interruption of power flow. Electrodes System 2 System System 1 System AC AC HVDC Overhead Line

Monopolar ground return

HVDC Ovoverhead Lineline Electrodes System 2 System System 1 System AC AC HVDC Ovoverhead Lineline

monopolarMonopolar metallic return

Fig. 5.6.3-1: HVDC system configurations

MRTB at Tian Guang/China DC switchyard

3234 5.6.3.3 Design Considerations The ground resistance Re is normally much With reference to Fig 5.6.3-4, the principle lower than the metallic resistance Rm. of commutation is as follows: At t0,the Details regarding the duties of ”HSNBS” Therefore, during the transitional steady- contacts of the breaker separate, thereby and ”HSGS” are not discussed here but state condition with both MRTB and GRTS introducing an arc into the circuit. The the more severe requirements for “MRTB“ closed, most of the current is flowing characteristic of this arc sets up an oscilla- and ”GRTS“ are explained. through ground which determines the tory current (frequency determined by commutation requirements for MRTB and Lp Cp) which is superimposed on the Fig. 5.6.3-2 shows the disposition of MRTB GRTS. I3 may reach values of up to 90% current I1. As Rp is very small, the oscillation and GRTS. Rm and Lm represent resistance of the total current I0 and I4 values of up is not damped but increases. As soon as and inductance of the transmission line. to 25% of I0. the current I1 passes through zero (refer Re and Le comprise resistances and The following considerations refer to MRTB to t1 in Fig. 5.6.3-4), the breaker current is inductances of the ground return path. only. From the above it can be concluded interrupted. I3 ,however, remains unchanged that the commutation duties for transfer now charging the capacitor Cp until it from ground to metallic return (MRTB) are reaches a voltage limited by the energy much heavier than from metallic to ground absorber. This voltage acts as a counter Rm Lm I4 return (GRTS). voltage to reduce the current I3 and to in- Fig. 5.6.3-3 shows the basic MRTB circuit. crease the current I4 (refer to Fig. 5.6.3-4 GRTS An energy absorber and the LpCp resonant and Fig. 5.6.3-3). When the absorber I3 Re Le circuit (Rp represents the ohmic resistance limiting voltage has been reached, the I0 of that branch only) are connected in current I3 flows into the absorber which MRTB parallel to the main switch (MRTB) which dissipates an amount of energy deter- is a conventional SF6-type AC breaker. mined by the counter voltage to bring I3 to zero. When I3 has dropped to zero, I4 equals I0 and the current commutation from ground to metallic return has been Fig. 5.6.3-2: Equivalent circuit relevant to completed. It should be noted that the MRTB and GRTS operation current I0 of the system (refer to Fig. 5.6.3-2) did not change, i.e. the power transmission was never interrupted. I R L 4 m m There are also MRTB principles other than Lp Cp the explained one which are based on GRTS Rp Lp Cp complex resonant circuits, externally exci- I3 Re Le ted with additional auxiliary power sources. I0 With respect to reliability and availability, MRTB the advantage of the above principle with I3 I1 Uarc Energy Absorber passive resonant circuit which is used by Siemens is quite evident. The nozzle system and specifically the flow of SF6 gas in the Siemens standard SF AC Fig. 5.6.3-3: Details of the MRTB circuit I1 6 breakers result in an arc characteristic t0 t1 t which ensures reliable operation of the tc passive resonant circuit. One unit of a Uabsorber standard three-phase AC breaker is used. Extensive series of laboratory tests have Uarc I shown the capabilities of Siemens SF6 3 breakers for this application. Furthermore, such switches are successfully in operation in various HVDC schemes.

Fig. 5.6.3-4: Principles of MRTB operation

3335 5.6.4 Earth Electrode

5.6.4.1 Function of the Earth 5.6.4.2.2 Vertical Land Electrode Electrode in the HVDC System If the ground strata near the surface have Earth electrodes are an essential Fill a high specific resistance, but component of the monopolar HVDC underneath, there is a conductive and transmission system, since they carry 1.5 – 2.5 m sufficiently thick stratum at a depth of several tens of meters, the vertical deep the operating current on a continuous Crushed Stone electrode is one possible solution. basis. They contribute decisively to the profitability of low-power HVDC systems, Conductor Figure 5.6.4-3 shows, as an example, (Iron) one of the four deep electrodes at Apollo, since the costs for a second conductor Coke Bed (with half the nominal voltage) are (0.5 x 0.5m2) the southern station of the Cahora Bassa significantly higher, even for transmission HVDC system. over short distances, than the costs for Fig. 5.6.4-1: Cross section through a horizontal Manhole the earth electrodes. land electrode Flexible Earth electrodes are also found in all Connection bipolar HVDC systems and in HVDC As shown in figure 5.6.4-1, the electrode Feed Cable multi-point systems. As in any high- conductor itself, which is generally made Concrete voltage system, the power circuit of the of iron, is laid horizontally at a depth of Cover approximately 2 m. It is embedded in HVDC system requires a reference point coke which fills a trench having a cross for the definition of the system voltage section of approximately 0.5 x 0.5 m2. as the basis for the insulation coordination Graphite Rod

and overvoltage protection. In a bipolar The advantage of this design becomes 80 m apparent in anodic operation. The passage Crushed HVDC system, it would conceivably be Stone possible to connect the station neutral of the current from the electrode conductor into the coke bed is carried Borehole ø:

point to the ground mat of the HVDC m primarily by electrons, and is thus not 10 0.57 m station to which the line-side star points associated with loss of the material. Coarse Grain of the converter transformers are also Graphite connected. But since the direct currents Several typical patterns of horizontal Conductive 40 m Layer in the two poles of the HVDC are never land electrodes are illustrated in absolutely equal, in spite of current figure 5.6.4-2 balancing control, a differential current flows continuously from the station Fig. 5.6.4-3: Vertical electrode at Apollo, the neutral point to ground. It is common a) b) Southern Cahora Bassa HVDC station practice to locate the grounding of the station neutral point at some distance (10 to 50 kilometres) from the HVDC c) d) e) f) station by means of special earth a) Line Electrode electrodes. b) Multi-Line Electrode c) Ring-Shaped Electrode 5.6.4.2 Design of Earth Electrodes d) Ring-Shaped Electrode with second ring Earth electrodes for HVDC systems may e) Star-Shaped Electrode be land, coastal or submarine electrodes. f) Forked Star Electrode In monopolar HVDC systems, which exist Fig. 5.6.4-2: Plan view of a typical design of almost exclusively in the form of sub- horizontal land electrodes marine cable transmission systems, there are fundamental differences between the design of anode and cathode electrodes.

5.6.4.2.1 The Horizontal Land Electrode If a sufficiently large area of flat land with relatively homogeneous ground characteristics is available, the horizontal ground electrode is the most economical form of a land electrode.

36 5.6.4.2.3 Cathodic Submarine Electrodes The design and construction of the Electrode cathodic submarine electrodes of a Module monopolar HVDC system with submarine power transmission cable do not present any particular problems. Since Feed there is no material corrosion, a copper Cable cable laid on the bottom should theo- retically suffice. The length of the cable Coke Bed Graphite Rod must be designed so that the current Concrete Cover density on its surface causes an electrical field of < 3 V/m in the surrounding water, which is also safe for swimmers and divers. 0.5 – 1 m Cable

5.6.4.2.4 Anodic Submarine Electrodes Figure 5.6.4-4 shows an example of a 2 – 5 m 0.5 – 1 m linear submarine electrode for anodic operation. The prefabricated electrode modules are lowered to the ocean floor Fig. 5.6.4.-4: Linear submarine electrode and then connected to the feed cable. (anodic operation) When the submarine electrodes are divided into sections which are con- 5.6.4.2.5 Anodic Coastal Electrode nected to the HVDC station by means of separate feed cables, the electrode The conventional design of a coastal can be monitored from the land. electrode is similar to that of a vertical land electrode. Graphite rods surrounded by a coke bed are installed in boreholes which are sunk along the coastline. The advantage of the coastal electrodes is easy accessibility for inspection, maintenance and regeneration, if necessary. A coastal electrode can also be configured in the form of a horizontal land electrode if the ground has the necessary conductivity or if the necessary conductivity can be achieved by irrigating the trench with salt water. In either case, it is assumed that even with a coastal electrode, the current flow to the opposite electrode takes place almost exclusively through the water.

37 5.7 Control & Protection

5.7.1 General The control system plays an important The Control and Protection System is role in the successful implementation of based on Siemens standard products. a high-voltage DC current transmission. This makes sure that many years of Reliability through redundant and fault- experience and thousands of applications tolerant design, flexibility through choice contribute to today's control and pro- from optional control centres and high tection development. At the same time dynamic performance were the prere- the system performance has been in- quisites for the development of our control creased substantially by the use of latest and protection system. Continuous technology CPUs. feedback during 30 years of operational experience and parallel use of similar The control is divided into the following technology in related applications yielded hierarchical levels: the sophisticated technology we can • Operator control system offer today. • Control and protection systems level •Field level (I/Os, time tagging, Main objectives for the implementation interlocking) of the HVDC control system are reliable energy transmission which operates highly efficient and flexible energy flow that responds to sudden changes in demand thus contributing to network stability.

Fig. 5.7.1-1: HVDC control hierarchy, one station (bipolar HVDC transmission scheme, redundancy is not shown in this figure)

GPS

Operator Control Level Master AC/DC Remote Interstation Workstations Clock I&M Control Interface Telecom

Local Area Network

Control Level Pole 1 Pole 2 Digital Station Control Control Control Controls Telecontr. & VBE & VBE Telecontr.

Field Bus

Optical Fibre

Process Level Measured Measured Measured Values AC Filter AC Feeder Transformer 1 DC Yard Values Valves Values Transformer 2 DC Yard

38 In the following, functions, tasks and components are described to provide an overview.

5.7.1.1 High Availability The main design criteria for Siemens HVDC systems is to achieve maximum energy availability. That as well applies Internet to the design of the control and pro- Internet tection systems. A single fault of any piece of equipment in the control and protection systems may not lead to a loss of power. Therefore, the major control and protection components are configured as redundant systems.

5.7.1.2 Self-Testing Features All control and protection systems are equipped with self-diagnostic features that allow the operator to quickly identify and replace the defective part to recover 5.7.1.5 Modular Design Fig. 5.7.1-2: Remote access connection redundancy as soon as possible. The control and protection systems use multiprocessor hardware. This means 5.7.1.3 Low Maintenance that the computing capacity can be With today’s digital systems there is no scaled according to the needs. requirement for routine maintenance. Therefore, the most economic solution However, should it be necessary to can be found in the first place, while replace single modules, the design is additional computing capacity can be such, that there is no operational impact added at any time later, should it be on the HVDC system. This is achieved needed. by designing all major components as redundant systems, where one channel 5.7.1.6 Communication Interfaces can be switched off without impact on the other channel. The control and protection systems as well as the operator control system 5.7.1.4 Best Support – Remote Access communicate via either Ethernet or Profibus protocols. For remote control As an optional feature, the control system interface, a number of major protocols can be accessed from the remote via is available. Other protocols can be the Internet. This allows plant monitoring, implemented as an option. detection of faults and from remote locations. To ensure the data security, a VPN (Virtual Private Network) encrypted connection is used. Furthermore, a password protected access makes sure that only authorized personnel have access. With the use of a standard web browser, main diagnosis data can be monitored. Expert access to the control components is also possible. This remote access feature provides excellent support for the commissioning and maintenance personell by our design engineers.

39 5.7 Control & Protection

5.7.2 Control Components Pole 1 Energy Transfer Control Location Control Level Runbacks Power Swing Pole 1 Mode Pole 1 DC-Sequences Stabilization 5.7.2.1 Operator Control System DEBLOCKED P=const. DC - WS STATION ENABLED ENABLED DEBLOCKED Pole 2 Energy Transfer Control Location System Configuration Runups Power Swing Pole 2 Mode Pole 2 DC-Power/Current Damping The tasks of a modern operations and DEBLOCKED P=const. DC - WS BIPOLAR ENABLED ENABLED DEBLOCKED monitoring system within the HVDC System Overview Date 03 - 16 - 94 Bipole Control Time control system include the following: 13 : 42 : 17 •Status information: clear and Station A Station B structured overview of the system for Ud = 500 kV Ud = 470 kV Pact = 900 MW Pact = 900 MW the operator Pmax = 1300 MW Pmax = 1300 MW α =15 Iact = 1800 A Iact = 1800 A γ =18 • Operator guidance: prevent Iset = 1800 A BP Pact = 1800 MW Iset = 1800 A Iramp = 500 A/min BP Pset = 1800 MW Iramp = 500 A/min maloperation, explain conditions Imax = 1980 ABP Pramp = 500 MW/min Imax = 1980 A I= 0A • Monitoring of the entire installation I= 0A and auxiliary equipment •Graphic display providing structural Pact = 900 MW α Pact = 900 MW =15 γ =18 overview of the entire system Pmax = 1300 MW Pmax = 1300 MW Iact = 1800 A Iact = 1800 A • Support of operating personnel through Iset = 1800 A Iset = 1800 A Iramp = 500 A/min Iramp = 500 A/min integrated operator guidance Imax = 1980 A Imax = 1980 A Ud = 500 kV Ud = 470 kV

F1 Control Level F2 Energy Transfer Mode F3 BP Power Setting Values Fig. 5.7.1-3: Operator workstation, typical F4 Power Direction F5 Current Setting Values F6 Reduced Voltage screen layout for a bipolar HVDC system F7 Bipole Block F8 Modulation Enable/Disable F9 Emergency Stop overview F10 Power Ramp Stop F11 Operator Notes F12 MENU

•Troubleshooting: support operator EVENT SELECTION SORTING HVDC Transmission System EVENTS JUMPERED 05-03-94 with clear messages that allow quick TRIP+WARN+STAT TIME EVENT RECORDING AND ALARM SYSTEM 0EVENTS 16:40:12 DISPLAY SELECTION DISPLAY RANGE ACKNOWLEDGE resume of operation ONLINE ACTUAL DISPLAY FROM 1100 TO 1118 OF 1118 EVENTS NUMBER OF ALARMS: 1 • Display and sorting of events time DATE TIME NR. GROUP EVENT DEVICE DESCRIPTION STATUS

tagged via global positioning system 05-11-93 16:38:42:4220396 12 S =XJ12 OLC: START SEQUENCERUN (GPS), automatic generation of 05-11-93 16:38:42:4230400 12 s =XJ12 olc: status standby - 05-11-93 16:38:42:7270949 27S =JS11 CA01Q0 BREAKER COMMAND ON process data reports 05-11-93 16:38:42:8620946 27S =XS11 CA01Q0 BREAKER ON ON • Display and storage of messages 05-11-93 16:38:42:9410949 27s =XS11 cac01q0 breaker command off 05-11-93 16:38:43:5790769 20WOLC:C-BANKOFFBLOCKED =XJ11 + concerning events, alarms and 05-11-93 16:39:05:082039712 S =XJ12 OLC: STATUS BLOCKED + disturbances 05-11-93 16:39:05:1280396 12 s =xj12 olc: start sequence stop 05-11-93 16:39:42:0600396 12 S =XJ12 OLC: START SEQUENCERUN 05-11-93 16:39:42:0610397 12 s =xj12 olc: status blocked - 05-11-93 16:39:42:2270806 21 W =XX11 TELE: H1 MESSAGE TRANS FAULT + 05-11-93 16:39:43:5590806 21 w =xx11 tele: h1 message trans fault - 05-11-93 16:39:52:4700981 29S =JS11 CA03Q0 BREAKER COMMAND ON 05-11-93 16:39:52:5560978 29S =XS11 CA03Q0 BREAKER ON 05-.11-93 16:39:52:5610301 09 s =v05 clc: converter blocked - 05-11-93 16:39:52:5610300 09S =V05 CLC: CONVERTER DEBLOCKED + 05-11-93 16:39:52:5960781 20w =XJ11 ca03q0 breakercommand off 05-11-93 16:39:52:9790981 29s =js11 OLC: Q-BAND UNDERFLOW +

05-11-93 16:39:53:021039812S =XJ12 OLC: STATUS DEBLOCKED + 05-11-93 16:39:53:0640396 12 s =XJ12 olc: start sequence stop 05-11-93 16:39:55:7030781 20 s =XJ12 olc: q-band underflow -

F1 PAGE UP F2 PAGE DOWN F3 ACKNOWLEDGE Fig. 5.7.1-4: Sequence of events recording F4 EVENT SELECTION F5 SORT SELECTION F6 PRINTOUT SELECTION (SER), screen layout for display of SER F7 EVENT ARCHIVE F8 JUMPER EVENT F9 REMOVE EVENT information F10 INITIATE GEN STATUS F11 RETURN TO PRESEN T CONDITIONS F12MENU

40 • Analysis of operating mode based HVDC Date: 18.08.1999 on user-defined and archive data Trend Display 13:11:56 (trend system)

Pole 1

Pole 2

F1: DISPLAY F2: PARAMETER SET F3: PRINT CONFIGURATION CONFIGURATION Fig. 5.7.1-5: Trend system, F4: SELECT F5: ARCHIVE F6: ARCHIVE example for different F7: F8: FREEZE F9: FREEZE trend displays F10: DEFAULT F11: ONLINE F12: MENU

5.7.2.2 Control and Protection Systems Level The major tasks in this level are: duplicated sensor head redundant channel • Measuring of actual values Shunt •Transmission of required amount of power Analog / Digital •Protection of equipment and personnel Optical Digital control/ protection Signal fibre system Measuring of Actual Values Digital / Optical Digital To determine the actual values, the hybrid optical DC measuring system is used. Id Electrical Energy Optical Energy Power fibre This system measures the voltage drop on a shunt in the high-voltage busbar, Optical Energy Electrical Energy Power supply converts this voltage into a light telegram Sensor Head and transfers it to the controls via fibre optics. Sensor Head Box at high voltage level Measuring at control systems

The scheme is designed completely Fibre optic cable redundant, therefore loss of a signal does not lead to an impact on power transmission. This measuring principle The even more obvious advantages of Fig. 5.7.1-6: Principle of the hybrid optical greatly contributes to an increased such a scheme: measuring scheme availability of the control and protection •Reduced weight (100 kg) scheme. •Passive system (linear response) •Reduced EMC (due to fibre optics) • Integrated harmonic measurement (Rogowsky coil) for use in active filters or harmonic monitoring schemes.

41 5.7 Control & Protection

Power Transmission Protection of Equipment and Personnel Some control actions are initiated by the protection scheme via signals to the The pole control system is responsible The DC protection system has the task control systems. With this monitoring for firing the thyristor valves such that to protect equipment and personnel system, a false trip due to a hardware the requested power is transmitted. also on a per pole basis.The protection fault of the protection hardware itself is The pole controls on each side of the systems can be divided into two areas. almost impossible. transmission link therefore have to fulfil The HVDC-related protection functions different tasks. The pole control system are referred to as DC protection. These The required functions of the various on the rectifier side mainly controls the include the classic DC protection functio- protective relays are executed reliably current such that the requested power nality consisting of converter protection, for all operating conditions. The selected is achieved, whereas the pole control DC busbar protection, DC filter protec- protective systems ensure that all possi- system on the inverter side controls the tion, electrode line protection and DC ble faults are detected, selectively cleared DC voltage such that minimum losses line protection as well as the AC filter and annunciated. The continuously active are achieved. protection and converter transformer self-monitoring system takes care that protection. defects of the DC-protection Hardware The pole control is configured as will be detected. redundant scheme, so that a failure in The AC protection scheme consists one channel has no impact on power mainly of the AC busbar, the AC line transmission. and the AC grid transformer protection. This channel will be repaired while the The task of the protective equipment is other channel stays in operation. Then to prevent damage of individual compo- the repaired channel will resume opera- nents caused by faults or overstresses. tion. In bipolar schemes, a redundant pole control system is – of course – Each protection zone is covered by at assigned to each pole. Failures in one least two independent protective units pole will not have any impact on the – the primary protective unit and the remaining pole. secondary (or back-up) protective unit.

Redundant Local Area Network 1 AC-Busbar Protection 2 AC-Line Protection Red. Field Bus 3 AC-Filter Protection 4 Converter Transformer Protection 5 Converter Protection Pole Red. Pole 6 DC-Busbar Protection System Control Control 3 7 DC-Filter Protection Selection 8 Electrode Line Protection Control stem 1 stem 2 9 DC-Line Protection

Sy Converter Converter Sy Control Control

9 Valve Valve Base Base 7 6 Electronic stem 1 stem 2 Electronic Sy Sy 8

4 Fibre Optic 5 Thyristor 1 Electronic

12-Pulse Group

Fig. 5.7.1-7: Redundant pole control system Fig. 5.7.1-8: Protection zones, one pole/one structure (for one 12-pulse group) station 42 All protective equipment in the HVDC converter station is realized either with the digital multi-microprocessor system or with digital Siemens standard protective relays. ”The DC Protection is of a fully redundant design. Additionally both protection systems incorporate main- and back-up protection functions using different principles. The AC Protection consists of a main and back-up system using different principles. The different Protection Systems are using different measuring devices and power supplies.”

5.7.3 Control Aspects 5.7.3.1 Redundancy All control and protection systems that contribute to the energy availability are configured in a redundant setup. This allows to cover any single fault in the control and protection equipment without loss of power.

5.7.3.2 Operator Training For Siemens HVDC application, an operator training simulator is available as an Fig. 5.7.1-9: Real-time simulator option. The simulator allows the operator to work with the same hardware and 5.7.4.2 Dynamic Performance Test software than in the real process. This simulator consists of the original operator The offline simulation with EMTDC is workstation plus a simulator PC, then already an extremely accurate forecast runs the HVDC process and feeds the of the real system behaviour. To verify relevant data to the workstations. the findings and optimize the controller settings, the control and protection 5.7.4 Testing and Quality Assurance systems are in addition tested during the dynamic performance test against a The design process has a number of real-time simulator. During that phase, defined review steps that allow to verify the customer may witness these per- the functionality and performance of the formance tests of the final control and controls before they are delivered on site. protection software. Already along with the tender, the use of accurate simulation tools allows to answer specific performance issues that are vital to the customer’s grid.

5.7.4.1 Offline Simulation EMTDC Siemens uses a simulation tool that includes a complete control library, with all details of control and protection functionality modelled in such detail that forecast of real system behaviour is 100% reliable. Therefore it is possible to optimize the application to find the best economic solution while providing the optimum performance.

43 5.7.4.3 Functional Performance Test This is required to assure the systems being free of transportation damage. In the functional performance test, the The next station-related tests are the delivered control and protection hard- subsystem tests. Subsystems consist ware is installed and tested against a of equipment items which are grouped real-time simulator. The purpose of the together according to common FPT is the proper signal exchange functions like AC filter banks or thyristor between the various control components valve systems. The main task is testing as well as the verification of the specified the proper function of interconnected control sequences. This allows optimized systems before switching on high commissioning time. Furthermore, voltage. Following this, station tests customer personnel can participate in under high voltage but no energy this test for operator training and to transfer will take place. Finally, system become familiar with the control system. Fig. 5.7.1-11: Example for a functional and acceptance tests with several performance test setup operating points of energy transfer will 5.7.4.4 On-Site Tests be used for final fine tuning and On-site tests are basically divided into At the precommissioning stage, the base verification of system performance. test steps regarding the related station work for commissioning the control (station A, station B) and into the test system and protection system is re- steps related to the whole HVDC system quired. The main task is preparation and (see figure 5.7.1-10). individual testing of any single system.

Hardware Design Cubicle Manufact. Bid Software Functional On-Site Stage Contract Design Spec. Design Perf. Tests Tests Operation Dynamic Perf. Test (TNA) EMTDC C&P Study

Fig. 5.7.1-10: The main steps for the HVDC control and protection versus the time starting from the contract award up to the on-site tests

44 System Studies, Digital Models, Design Specifications 6

6.1 System Studies 6.2 Digital Models

During the planning stage of a HVDC project preliminary Digital models of HVDC system can be developed studies are carried out in order to establish the basic according to the specified requirements. Typically a design of the whole HVDC transmission project. This digital model of dc system is needed for a specific load includes the co-ordination of all relevant technical parts flow and stability simulation program, while another of the transmission system like HVDC converters, AC digital model is required for simulation in a typical and DC overhead lines as well as the submarine cable electromagnetic transients program such as EMTDC. if applicable. The functionality and settings of HVDC control and All specified requirements will be taken into account protection system shall be represented in a proper and are the basis for the preliminary design of the HVDC manner in such models, which allow suitable simulation transmission link. In addition, special attention is paid of steady state and transient behavior of HVDC system to improve the stability of both connected AC systems. in the corresponding digital programs. Digital models Several additional control functions like power modulation, consistent with the actual dc control and protection frequency control, AC voltage limiter can be included in system are beneficial both for the operation of the HVDC order to provide excellent dynamic behaviour and to scheme and for the network studies including DC link. assist the AC systems if the studies show it necessary. Typically such models can be developed on request in Sub-synchronous oscillation will be avoided by special the detailed project design stage when all major design control functions if required. All the AC system conditions works of control and protection functions are completed. and the environmental conditions as given in the relevant documents will be considered in the design calculations. 6.3 Control and Protection Design Specifications The final design of the HVDC transmission system including the operation characteristics will be defined Design Specifications are written for the control, during the detailed system studies. All necessary studies protection and communication hardware and software. are carried out to confirm the appropriate performance The control panels are then designed, manufactured requirements and ratings of all the equipment. inspected and tested in accordance to the design Due consideration is given to the interaction with the specification. The software for the control and protection AC systems on both sides, the generation of reactive is also written in accordance to the design specification. power, system frequency variations, overvoltages, short It is tested using real time simulators in the dynamic circuit levels and system inertia during all system performance test and functional performance test. configurations. Specifications for the topics below are typically written: Typically the following studies are carried out: a) General Control and Protection a) Main Circuit Parameters b) Interface Systems c) Station Control b) Power Circuit Arrangements d) Diagnosis Systems c) Thermal Rating of Key Equipment e) Pole Control d) Reactive Power Management f) HVDC Protection e) Temporary Overvoltages and Ferro-resonance g) AC Protection Overvoltages h) Metering and Measuring f) Overvoltage Protection and Insulation Coordination i) Operator Control g) Transient Current Requirements j) Communication h) AC Filter Performance and Rating i) DC Filter Performance and Rating j) AC Breaker and DC High-Speed Switch Requirements k) Electromagnetic Interference l) Reliability and Availability calculations m) Loss Calculation n) Subsynchronous Resonance o) Load Flow, Stability and Interaction between different HVDC Systems p) Audible Noise

45 7 Project Management

7. 1 Project Management in HVDC Projects Project Management The success and functional completion Quality Procurement Assurance Technical Commercial of large projects depends on the struc- Project Manager Project Manager turing of the project team in accordance to the related work and manpower Control & coordination. Periodically updates and Basic Design Station Design Protection Logistic adaptation of design guarantee the Engineering Documentation execution of the project with constant Training high quality within the target time frame. Transient Throughout all production, working Civil Network Engineering Analyser process and on-site activities, health, Functional Commercial safety and environmental protection Main Performance Components Civil (HSE) measures s well as application of Construction Test Commissioning commonly agreed quality standards such Installation as DIN EN ISO 9001 are of prime Communication Financing Systems importance to Siemens.

7. 1.1 Division Responsibilities Fig.7-1: Project organisation plan The overall project is divided and organised according to design activities 7. 1. 3 Risk reduction 7. 1. 5 Scheduling and technical component groups. These features make it possible to define clear Any risks that could arise due to incorrect The hierarchically structured bar-chart function packages which are to deadlines, unclear technical concepts or schedule is a high-level control tool in a great extent homogeneous within excessive costs shall be recognised early project management. The clear structure themselves and can be processed with enough by a monitoring system so that of sequential processes and parallel minimised interfaces. counter measures can be taken. This activities is crucial for execution of a increases contract quality and creates 24 to 36 month duration, according to 7. 1. 2 Tr ansparency the basis for clear design criteria. the project requirements.

A clear process structure plan (PSP) 7. 1. 4 Progress Report Deadlines for project decisions – standardised for HVDC projects makes especially those of the critical path – can the project contents and sequences Periodically meetings with subcon- easily be identified enabling the project transparent in their commercial and tractors, in house control working teams manager to make up-to date pre-esti- technical aspects. Associations and and customer are recorded in progress mates and initiate suitable measures in interactions are clarified according to reports which form an integral part of due time. procedure of work. the quality insurance system.

Activity Time

Award of Contract Engineering/System Studies Manufacturing Transportation Civil Works & Buildings Erection & Precommissioning Station Tests System Tests Commercial Operation Fig.7-2: Structured bar-chart timeschedule

46 47

47 Siemens AG Power Transmission and Distribution High Voltage Division Postfach 32 20 91050 Erlangen Subject to change without prior notice Order No. E50001-U131-A92-V2-7600 Germany Printed in Germany Dispo-Stelle 30000 www.siemens.com/hvdc 61D7085 TV/EV 02032.