PSO with segmented mutation for site selection in grid-connected photovoltaic power generation system

This paper presents a novel Segmented Mutation Particle Swarm Optimization (SMPSO) algorithm to address the selection of photovoltaic (PV) array sites and electrical transformer sites in the planning phase of grid-connected PV power generation systems. The site selection process for PV arrays and electrical transformers directly affects both the system's power generation efficiency and costs. However, this site selection task poses significant challenges for optimization algorithms. PSO is one of the most widely used population-based optimizers with a wide range of applications. Due to the shortcomings of premature convergence and local optima entrapment of PSO on site selection problem, this paper gives a piecewise variation PSO, which adopted global particle variation operation in the early stage of iteration to improve the global search ability, and used constrained variation and small-scale variation operation of better particles in the later stage of iteration. Through comparison with optimization results obtained from standard test functions, it is confirmed that this algorithm possesses superior performance in optimizing multimodal functions. Additionally, the designed algorithm is applied to effectively resolve the PV array site selection and electrical transformer site selection problems. Empirical findings derived from experiments demonstrate that the proposed SMPSO algorithm surpasses alternative PSO algorithms in terms of solution quality and convergence speed. Its implementation is expected to yield cost-efficient designs for photovoltaic power generation systems.