OPTIMAL INTEGRATION OF STEAM TURBINES IN INDUSTRIAL PROCESS PLANTS J D Kumana, MS ChE Kumana & Associates, Houston, Tx [email protected] (281) 437-5906 AIChE-STS, network meeting Houston, 2 Feb 2018 Outline • Definitions: CHP and Efficiency • Thermodynamics Review • Energy integration theory • CHP models Kumana & Associates © 2018 2 DEFINITIONS: CHP & EFFICIENCY • CHP = Combined Heat and Power (= energy utility system for the plant site) • Steam Turbines are Heat Engines that operate on the Rankine cycle. They convert DP into Shaftwork; a generator then converts Shaftwork into Elec power • Thermodynamic Efficiency is defined as Useful Energy Output Energy Input • For Generation, 1 useful output = Power only. Machine eff = ~20%, System Eff = ~35% • For Cogeneration, 2 useful outputs = Power + Process Heat, Machine eff = ~20%, but System Eff ~75-80% Kumana & Associates © 2018 3 This is CHP, but not Cogeneration LATENT HEAT OF ST EXHAUST IS WASTED BOILERS FUEL STEAM PROCESS STEAM KW TURBINE Pure power gen CONDENSER BOILER EFF ~ 80% POWER GEN EFF < 25% Kumana & Associates © 2018 4 This is both CHP and “Co-Generation” LAT HT OF EXHAUST STM IS USED IN THE PROCESS BOILERS FUEL PROCESS KW STEAM TURBINE LP STEAM OVERALL EFF ~ 75% Kumana & Associates © 2018 5 Alternative Cogen configurations Extraction Turbine Induction Turbine Kumana & Associates © 2018 6 Variations – hybrid Cogen and Condensing Extraction turbine Induction turbine Kumana & Associates © 2018 7 Simple Rankine Cycle flowsheet Schematic shown is for cogeneration mode Kumana & Associates © 2018 8 Difficult to match Heat:Power ratio of process Most efficient 4 Basic Configs – which do you think is most efficient? Kumana & Associates © 2018 9 The ultimate Combined-cycle Cogen scheme EXHAUST TO ATMOS HP STEAM HRSG LP STEAM PROCESS GAS TURBINE KW KW ELEC EXPORT AIR GAS OVERALL EFF ~ 85% Kumana & Associates © 2018 10 Different types of ST Efficiency Machine Efficiency = M, P1, T1, H1 • W/Qin = (H1-H2)/H1 • Isentropic Efficiency W = W/[M.(H1–H2)max] = (H1-H2)/(H1-H’2) M, P2, T2, H2 m M - m • System efficiency 3413.kW (M - m).2 m.H 2 PROCESS M.H1 Equipt H’2 = exhaust vapor enthalpy IF the expansion were isentropic (which it is not, and can never be) Kumana & Associates © 2018 11 A Bit of History … US Power plants stopped cogenerating ~1960 Kumana & Associates © 2018 12 THERMODYNAMICS REVIEW Rankine cycle on the P-V diagram P-V Diagram for Water 3500 Rankine Cycle 3000 1. BFW pumpimg 2. Stm gen in Boiler 2500 3. ST expansion + pow er gen 2 2000 4. Condensation 1500 Pressure,psia 1 1000 3 500 4 0 0.01 0.1 1 10 Specific Volume, ft3/lb Kumana & Associates © 2018 13 Power generation step (#3) on Mollier Chart Mollier Chart (H-S) for Steam 1600 1200 600 1550 Saturation 300 Pr, psig 150 1500 50 Quality, % stm 1450 Temp, F • Adiabatic 1.0 1400 1 800 expansion 600 (from 600 psig, 1350 700oF to 50 psig) 1300 500 700 1250 400 300 • Isentropic 1200 2 Enthalpy,Btu/lb 215 1150 efficiency Saturation 1100 line 1050 97 % 1000 92% Quality --12 psig 950 900 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 Kumana & Associates © 2018 Entropy, Btu/lb-F 14 Effect of P2/P1 on Machine Efficiency (W/Qin) Near-optimal Power-to-Heat Ratio vs Steam Pressure Ratio 0.25 Inlet Conditions for industrial cogen systems 0.20 Condensing Turbine limit 0.15 BPST 0.10 limit Power:Heat ratio Power:Heat 0.05 0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 P2/P1 Theoretical Machine Efficiency tops out at ~13% for BPST and 24% for CST before moisture content in turbine reaches dangerous levels. Kumana & Associates © 2018 15 Effect of P2/P1 on System Efficiency System Efficiency vs P2/P1 ratio 100% 80% 60% Condensation starts at P2 = 53 psig 40% System energy System Eff Exhaust Exhaust stm is dry 20% stm is wet 0% 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 P2/P1 System Efficiency peaks when exhaust steam is saturated, drops rapidly as P2/P1 is falls, slowly as P2/P1 rises Kumana & Associates © 2018 16 Next: What is the Optimum Exhaust Pressure? • P2 should be at a high enough pressure that it can be used for process heating • If there are multiple steam levels in the process, an extraction type turbine should be considered, with both exhaust pressures above ambient. • The amounts should match the process steam requirements ( “thermal match”) • For higher P2 or W/Qin increase P1 and T1 PINCH ANALYIS provides the ANSWER Kumana & Associates © 2018 17 OPTIMUM TURBINE INTEGRATION Qhot Pinch = minimum DT reqd for ht tr “Pinch Analysis” Temp Qhot & Qcold are the energy targets Qcold Heat Load It is possible to consolidate ALL the heating and cooling duties in the process into two Composite Curves that show the enthalpy change requirements between the entire temperature range over which the process operates Kumana & Associates © 2018 18 The Pinch Principle - 1 T QHmin+ XP Source XP Sink ? Not economic because DT< DTmin QCmin+ XP H If we allow XP heat transfer, Qh and Qc both increase by XP Kumana & Associates © 2018 19 The Pinch Principle - 2 To achieve the Energy Targets, DO NOT T Cooling • use Steam below Pinch Water Pinch Process Heat • use CW above Pinch Temp Transfer • transfer heat from process streams above Steam Pinch to process streams below Pinch Kumana & Associates © 2018 20 Steam Turbine Integration options A + W A + Q T T A - (Q - W) Q A Q Heat Engine W Q - W Heat Engine W Q Q - W Heat Engine W Q - W B + (Q - W) B - Q No improvement in system 100% conversion of Q W Kumana & Associates © 2018 21 Summary of Energy Balances = Machine efficiency Kumana & Associates © 2018 22 Grand Composite Curve - GCC HP STEAM T LP STEAM Used for utilities selection COOLING WATER REFRIGERATION H Kumana & Associates © 2018 23 Correct Integration of Steam Turbine Q • GCC shows us T LOSS exactly how much HP and LP steam Fuel is needed, and the W right P/T levels HPS • ST must always exhaust ABOVE LPS Grand Composite the Process Pinch Curve Process • When designed Pinch this way, payback CW is very good, H typically 3-4 yrs Fuel = HPS + LPS + W + Qloss Kumana & Associates © 2018 24 Total Site Source-Sink curves T Net process Req. Q cooling demand = available heat Sink w HP Source Net process LP stm Sink heating demand BFW CW Enthalpy, MMBtu/h Kumana & Associates © 2018 25 Optimize Configuration Fuel + EXISTING HP HP + MP MP LP LP Sink Reduction in fuel consumption + Fuel HP HP + + MP Source MP OPTIMIZED LP IP LP Power generation increased LLP LLP Kumana & Associates © 2018 26 CHP SIMULATION MODELS Kumana & Associates © 2018 27 Excellent Tool for Analysis Model should include all Key System Features: . Multiple steam levels . Multiple boilers (with eff. curves) . Process WHBs . Steam and Gas turbines (incl HRSG) . PRVs, Desuperheaters . Condensate recovery (by steam pr level) . Boiler blowdown flash & HX . Deaerators (could be > 1) . “Dump condenser”, if needed . Economizer for BFW preheat . BFW integration with process . Process power demand Kumana & Associates © 2018 28 CHP Optimization Guidelines • Set BPST exhaust pressures based on process steam headers (from GCC) • Set steam flows through BPSTs based on process heating duties at each Pr level • Condensing Turbines invariably a BAD idea • Minimize flows through PRVs • Use highest feasible DeAerator pressure(s) • Maximize condensate recovery • Preheat cold BFW makeup water by using it as a cooling medium in the process Kumana & Associates © 2018 29 On-line Utilities Optimization Real-Time Optimizer finds the best way to operate all utilities subject to contractual, environmental and operational constraints Optimum Optimum Set Points Emissions Measurements Utilities Regulations Operations Report Utility Systems Hydrogen Fuel Steam Water Electricity Key Performance Indicators Monitoring External Utilities Industrial Process Contracts and Site Accounting Reports Kumana & Associates © 2018 From VisualMesa® brochure, Courtesy of Soteica LLC, Houston, Tx 30 Expected Benefits and Costs • Typical savings = 3-5% of baseline (operator- optimized) energy costs • Typical installed cost = $500-900K • Typical Payback << 1 yr • Proven in dozens of Oil refineries, Chemical plants, Pulp/Paper mills (can be deemed a Best Practice) Kumana & Associates © 2018 31 Case Study – Operate closer to Optimum Optimización Visual Mesa Ahorros Anuales Predichos - Siguiendo las Sugerencias de Optimización TODOS LOS DATOS RECOLECTADOS, HASTA EL PRESENTE 10.00 9.00 8.00 7.00 6.00 5,30 % 5.00 > 2 MM €/year 4.00 > 4% 3.00 Savings / Total Energy Costs (%) Costs Energy Total / Savings 2.00 1.00 1,19% - 2-1 9-1 6-2 6-3 16-1 23-1 30-1 13-2 20-2 27-2 26-12 19-12 Day Ahorros Anuales Promedio del Período Anterior a la Optimización Promedio Después de Tomar Acciones de Optimización Y axis = Deviation from Optimum = Remaining Savings Opportunity Kumana & Associates © 2018 Graphic supplied by Soteica LLC, Houston, Tx 32 IN CONCLUSION • Use GCC to choose Stm Levels and Loads • Use BPSTs in cogen mode when possible • Condensing steam turbines are Invariably Bad* • Use TSSS to identify optimum CHP structure • Use CHP models to optimize parameters • Always optimize process demand before trying to design/optimize the CHP system • Ability to export excess power to the Grid at a fair price is critical to optimizing energy efficiency at National scale, and minimizing global GHG emissions with a few rare exceptions Kumana & Associates © 2018 33 Optimum Process Integration It’s like a jig- saw puzzle, but well worth the effort Kumana & Associates © 2018 34.