Methodology using process integration for identifying suitable Organic Rankine Cycles for waste heat valorisation Matthias Bendig*, Prof. François Maréchal, Prof. Daniel Favrat Industrial Energy Systems Laboratory (LENI) École Polytechnique Fédérale de Lausanne (EPFL) *corresponding author: [email protected] World energy-related CO emission savings by policy measure in the 450 scenario of the IEA. Source:IEA WEO 2009. Context Energy efficiency is essential: CO -abatement Cost efficiency Problem The optimal integration of an electricity production cycle into a process. Objective Propose a methodology in order to identify a pareto-optimal ORC: quantitative criteria: efficiencies, cost, CO2-equivalents, ODP, LCA qualitative criteria: toxicity, flammability Acknowledgements The authors thank Nestlé Suisse SA for all the support and letting us “spook around” their factory. The research, leading to these results, has as well received support from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 256790 (‘LOVE'). References Methodology How to adapt the ORC to heat sources and sinks of a process? Which are the relevant Constraints? Many parameters: Borel, L. and Favrat, D., 2010, Thermodynamics and Energetics for Engineers. EPFL Press, Switzer- land. International Energy Agency, 2009, World Energy Outlook, volume 2009. OECD/IEA, Paris. Maréchal, F., 2008, Pinch analysis, Chapter 3.19.1.7, UNESCO Encyclopedia of Life Support Systems, EOLSS Publishers Co Ltd. Muller, D.C.A., Maréchal, F., Wolewinski, T., Roux, P.J., November 2007, An energy management method for the food industry, Applied Thermal Engineering, Volume 27, Issue 16, Pages 2677-2686. 2 2 >50% Working Fluids and Mixtures Single- or Multi-Stage, Extraction Optimal Size Industrial Process Model for Thermodynamics of Industrial Process Technology Models of Low Temperature Heat Valorisation (LTHV) Iterate until optima found. Data Studies Measurements Deduce energy require- ments and integrate with Pinch-Analysis Evaluation of Objectives Constraints The number of decision variables can be reduced by: Limiting CO2-equivalents/ GWP Limiting ODP Excluding flammability Excluding toxicity The decision variable range can be reduced by: Limiting maximum pressure Limiting number of ORC-stages Limiting maximum compo- nents in working fluid Thermodynamics: - Energy-Efficiency - Exergy-Efficiency Cost (relative and NPV) Global Warming Potential Ozone Depletion Potential Life Cycle Analysis Mutations Alleles from high performing “Parent Configu- rations” “Genes” fixing the Configuration Evolutionary Multiobjective Optimisation (MOO) Pareto-Optimal Solutions Objective A Objective B Different solution “families” repre- senting different Technologies Solutions are pareto-optimal for at least two objectives. 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 -50 0 50 100 150 200 250 300 Heat (Q) [kW] corrected Temperature (T*) [°C] cold MER 1600 kW hot MER 789.2 kW internal heat recovery 6150 kW current cold utilities 3600 kW current hot utility 2789.2 kW current internal heat recovery 4150 kW Residual Heat 1237 kW <− Pinch-Point hot streams T* cold streams T* real Temp Integrated Composite Curve with supercritical R-134a ORC CU2 2750 kW 13 230°C 1 20°C P1 Gas 0°C Valve Comp 150 kW CU3 750 kW E1 600 kW R1 R2 HU 2789.2 kW P2 F1 F2 100°C 120°C 250°C 190°C 132.9°C 156.6°C 40°C 150°C 56.7°C 200°C HEX1 2000 kW HEX2 900 kW HEX3 1250 kW 2 4 6 7 8 9 10 11 14 15 16 17 18 19 5 30°C 50°C R3 F3 20°C 20°C 12 CU1 250 kW 40°C 3 H2O NH3 R-134a R-1234yf CO2 R-11 Toluene Hexamethyldisiloxane R-245fa Fraction of working fluids and mixtures: x =X x =X (1-x ) x =X (1-x -x ) x =X (1-x -x -x ) x =1-x -x -x -x For N working fluids N-1 deci- sion variables X are necessary. 1 1 1 2 1 2 2 3 3 i-1 i-1 1 2 i-2 i 1 2 i-2 i-1 i Pump Expander Condensor Evaporator II ORC with extraction Two-Stage ORC Pump Expander Condensator Evaporator 1 2 3 4 Evaporator Pump Expander Condensator Evaporator Simple ORC A single decision variable X deciding on stages and type. Further decision variables decide on temperatures and pressures. ORC