Multi-objective approach for determination of optimal operating parameters in low-temperature ORC power plant

Abstract

Power plant with organic Rankine cycle (ORC) is currently the most perspective technology for electricity production when low-temperature source of energy is to be used. Because of the ‘narrow’ temperature range (ca 70 ÷ 200 K) between upper and bottoming energy sources of the ORC, it is characterized by a low thermal efficiency. Thus, a lot of research and scientific effort have gone into developing new calculation techniques to optimize the work of such a system and the selection of a working fluid. In most of the recent studies a multi-criteria approach was applied and a single optimal operating point of the ORC was indicated as the final result of the optimization. The primary goal of this paper is to present an alternative approach. The general idea is to determine and indicate a set of the optimal operating parameters for which none of the considered criteria dominates significantly. In order to meet these conditions, multi-objective optimization using weight function G and genetic algorithm (GA) is conducted. The outcomes of the proposed model converge with Pareto set findings obtained using commonly known Non-dominated Sorting Genetic Algorithm-II (NSGA-II) method. The analysis is carried out for three sampled working fluids of the ORC: R1234yf, R1234ze and R245fa. The final results show that R1234yf is the optimal choice, since, in case of that fluid, the lowest values of the function G are observed for all weight distribution cases. Specifically, when the economic criterion is dominant (w1 > w2 case), the function G for R1234yf, R1234ze and R245fa is equal to: 0.1069, 0.1270 and 0.1460, while for the environmental criterion predominance (w2 > w1 case), the following values of G are obtained: 0.0985, 0.1260 and 0.1474. Simultaneously, the component criteria of the function G, i.e. the APR parameter (economic criterion) and SI index (environmental criterion) are minimized as well for all considered cases. In accordance with the fluid order given above, the optimized values of the APR parameter for w1 > w2 case are as follows: 0.61 m2 kW−1, 0.68 m2 kW−1 and 0.68 m2 kW−1, while, for w2 > w1 relation, the SI index is equal to: 0.48, 0.49 and 0.48. Furthermore, using the optimized values of the decision variables (pinch point temperature difference, evaporation temperature and condensation temperature), the novel 3-D plots with so-called sustainable surface are developed in the study to reveal the potential optimal ranges of important ORC parameters, such as: thermal efficiency, exergy efficiency or net power output. In general, the method should be a useful tool in the design process of the ORC systems, since it allows for the straightforward indication of several optimal operating points with respect to the adopted criteria. In other words, the designer has a wider choice and can opt for the parameters values that favor the selected objective.