A Genetic Algorithm solution to the Unit Commitment problem based on real-coded chromosomes and Fuzzy Optimization
This paper presents a combined Genetic Algorithm - Fuzzy Optimization approach to the Unit Commitment problem. The Unit Commitment problem is a high complex combinatorial optimization task, nonlinear and large-scale. In order to obtain a near optimal solution in low computational time and storage requirements, with respect to all specified constraints, a Genetic Algorithm using real-coded chromosomes is proposed in opposite to the more commonly used binary coded scheme. Gathering data from a list of strict priority order the Genetic Algorithm generates different candidate solutions to the problem, whereas Fuzzy Optimization guides the whole search process under an uncertain environment (varying load demand, renewable energy sources). A system consisting of 10 generating units is presented to demonstrate application of the proposed algorithm to the Unit Commitment problem. The obtained results show satisfactory outcome in total cost, compared to Dynamic Programming based applications and the sole Genetic Algorithm based solution to the Unit Commitment problem.