Improved genetic algorithm for optimal multistage transmission system planning

This paper presents an improved GA approach to optimal multistage transmission network planning. A fitness function including investment and overload constraint is constructed. The overload is checked by DC load flow. A concise codification model called redundant binary coded technique is proposed. By this technique the crossover operation can be executed inside the gene so that the re-combinatorial and search function of the crossover operator are well utilized. The Simulated Annealing selector is used to adjust the fitness function in the evolution process. Some improvements are employed to speed up the algorithm convergence such as keeping excellent seeds, mutation in pair, etc. Based on the proposed model, a computational program has been developed. Three case studies are applied to demonstrate the usefulness and effectiveness of the suggested multistage transmission network planning model.