Customizable optimization of clean energy base subsystems in subtropical monsoon regions

Abstract

This paper introduces a novel, customizable multi-objective optimization framework and presents C-$ϵ$-EGO, a Bayesian-based global optimization algorithm designed to tackle constrained mixed-integer multi-objective programming problems. The multiple advantages of establishing a wind-solar-pumped-storage clean energy base in a subtropical monsoon climate are thoroughly demonstrated based on precipitation and water availability, wind and solar energy resources, and terrain suitability. By integrating detailed models of wind power subsystems, photovoltaic power subsystems, and pumped storage subsystems with real-world operational parameters, our approach enables a customizable optimization strategy for 100% clean energy bases in subtropical monsoon climates. The algorithm transforms traditional multi-objective problems into a constrained single-objective formulation using an enhanced epsilon constraint method and a penalty function approach, resulting in a uniformly distributed Pareto front. In our case study, 15 Pareto-optimal solutions are obtained that meet predefined numerical constraints, providing valuable practical reference points for engineering decision-making. Comparative analysis against 19 established multi-objective optimization algorithms demonstrates the superior performance of the proposed method, offering a robust tool for balancing economic and reliability objectives in the planning and deployment of integrated clean energy systems.