An efficient hybrid CS‐DP application for generation expansion planning problem

Abstract

This article proposes a new hybrid approach for generation expansion planning (GEP) based on cuckoo search (CS) and dynamic programming (DP). Generally, the GEP problem is known as having very large search space and highly nonlinear complex combinatorial optimization. The global optimal solution is to solve with the full enumeration DP. However, the computation time grows exponentially when the planning horizon and candidate options increase, which is called ‘curse of dimensionality’. To solve this problem, a hybrid CS‐DP approach is proposed with three techniques for solution quality and convergence characteristic enhancement. First, the CS is integrated into the DP structure for the optimum solution of the very large search space problem. Second, the feasible search concept can eliminate infeasible solutions from the searching process. Finally, the memory exploration search is applied to prevent repeated searches. To measure its effectiveness, the proposed method is applied to 15 existing power plants, with 5 candidate options and 2 test cases: 14 year and 20 year study periods. The experimental results are compared with classical and metaheuristic optimizations. The test results indicate that the proposed hybrid CS‐DP achieves higher solution quality and superior convergence characteristics than other methods. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.