The presented work compares two optimization models for combined heat and power (CHP) energy systems scheduling. Both models are focused on the evaluation of the best operating strategy to run a defined cogeneration system that is dealing with time-variable loads and tariffs. The simultaneous usage of different prime movers operating in parallel is taken into consideration as well as their part load performance, the influence of ambient temperature, and the usage of a heat storage system. One of the models is heuristic and adopts an optimization strategy based on a multi-step approach: it simulates several cases according to a pre-defined path exploring the most reasonable operational modes and comparing them systematically. The other one relies on a mathematical approach. A Mixed Integer Linear Programming (MILP) algorithm has been developed in order to deal with more complex systems without the need of predefining a too large variety of optimization paths according to the case. Results of the two models are compared against a test case based on real plant specifications, discussing their performance, by the point of view of simulation capabilities, quality of the optimization results, and calculation time.
Cogeneration systems optimization: Comparison of multi-step and mixed integer linear programming approaches
BISCHI, ALDO;CAMPANARI, STEFANO;MANZOLINI, GIAMPAOLO;MARTELLI, EMANUELE;SILVA, PAOLO;MACCHI, ENNIO;
2016-01-01
Abstract
The presented work compares two optimization models for combined heat and power (CHP) energy systems scheduling. Both models are focused on the evaluation of the best operating strategy to run a defined cogeneration system that is dealing with time-variable loads and tariffs. The simultaneous usage of different prime movers operating in parallel is taken into consideration as well as their part load performance, the influence of ambient temperature, and the usage of a heat storage system. One of the models is heuristic and adopts an optimization strategy based on a multi-step approach: it simulates several cases according to a pre-defined path exploring the most reasonable operational modes and comparing them systematically. The other one relies on a mathematical approach. A Mixed Integer Linear Programming (MILP) algorithm has been developed in order to deal with more complex systems without the need of predefining a too large variety of optimization paths according to the case. Results of the two models are compared against a test case based on real plant specifications, discussing their performance, by the point of view of simulation capabilities, quality of the optimization results, and calculation time.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
