The unit commitment (UC) problem is a nonlinear, high-dimensional, highlyconstrained, mixed-integer power system optimization problem and is generallysolved in the literature considering minimizing the system operation cost asthe only objective. However, due to increasing environmental concerns, therecent attention has shifted to incorporating emission in the problemformulation. In this paper, a multi-objective evolutionary algorithm based ondecomposition (MOEA/D) is proposed to solve the UC problem as a multi-objectiveoptimization problem considering minimizing cost and emission as the multipleobjec- tives. Since, UC problem is a mixed-integer optimization problemconsisting of binary UC variables and continuous power dispatch variables, anovel hybridization strategy is proposed within the framework of MOEA/D suchthat genetic algorithm (GA) evolves the binary variables while differentialevolution (DE) evolves the continuous variables. Further, a novel non-uniformweight vector distribution strategy is proposed and a parallel island modelbased on combination of MOEA/D with uniform and non-uniform weight vectordistribution strategy is implemented to enhance the performance of thepresented algorithm. Extensive case studies are presented on different testsystems and the effectiveness of the proposed hybridization strategy, thenon-uniform weight vector distribution strategy and parallel island model isverified through stringent simulated results. Further, exhaustive benchmarkingagainst the algorithms proposed in the literature is presented to demonstratethe superiority of the proposed algorithm in obtaining significantly betterconverged and uniformly distributed trade-off solutions.
展开▼
