An enhanced NSGA-II and a fast constraint repairing method for combined heat and power dynamic economic emission dispatch

Abstract

In this paper, an enhanced NSGA-II (ENSGA-II) is presented for the combined heat and power dynamic economic emission dispatch (CHPDEED). ENSGA-II produces offspring individuals by a novel crossover of the Cauchy distribution with adaptive location and scale parameters. It can not only sufficiently explore and exploit the decision space but also remain a relatively high population diversity. Also, ENSGA-II adopts a modified crowding distance to penalize crowded individuals, seeking to evenly distribute the individuals in the objective space. In addition, a fast constraint repairing method (FCRM) is proposed to significantly reduce the constraint violations and guide infeasible individuals to move towards feasible zones rapidly. The proposed ENSGA-II is applied to a number of multi-objective optimization problems and compared with the other eight multi-objective evolutionary algorithms (MOEAs). ENSGA-II is found to produce better results on most problems, such as smaller inverse generational distances, larger hypervolumes, larger coverage rates and smaller spacings. Therefore, ENSGA-II can obtain a satisfactory Pareto set with broad spread, high diversity, strong convergence and good evenness, and it is an efficient alternative for CHPDEED.