Modified Shuffled Frog-Leaping Algorithm on Optimal Planning for a Stand-Alone Photovoltaic System

Abstract

Shuffled frog leaping algorithm (SFLA) is a new meta-heuristic evolutionary algorithm with simple algorithm and effective calculation speed. It performs stochastic searching process that mimics natural biological evolution and the social behavior of species. SFLA conducts its formulation from two main methods, the local searching validated by particle swarm optimization and the competitiveness mixing of information implemented by shuffled complex algorithm. A modified shuffled frog leaping algorithm (MSFLA) is investigated that improves the leaping rule by properly extending the leaping step size and adding a leaping inertia component to account for social behavior. In this paper, a MSFLA is proposed to solve combinatorial optimization problem for a stand-alone photovoltaic (SPV) generation planning. Several practical installed and operated costs for a SPV system were used to trade off nonlinear system reliability in the three specified locations of Taiwan. Different degrees of loss of load hours and load profiles were investigated to achieve the long-term planning requirement.