Comprehensive learning Jaya algorithm for parameter extraction of photovoltaic models

摘要

Given strong global search ability and less sensitive to initial solutions, many metaheuristic algorithms have been successful used to extract the unknown parameters of photovoltaic (PV) models. However, most applied metaheuristic algorithms need extra control parameters except the essential population size and terminal condition. For unknown optimization problems, how to set these control parameters to get the optimal solutions is a great challenge. To overcome this challenge, this paper presents a novel metaheuristic algorithm called comprehensive learning Jaya algorithm (CLJAYA) for parameter extraction of PV models. CLJAYA is a new variant of Jaya algorithm, which enhances global search ability of Jaya algorithm by the designed comprehensive learning mechanism. CLJAYA has a simple structure and only needs the essential population size and terminal condition for optimization. To verify the effectiveness of the improved strategies, CLJAYA is first employed to solve the well-known CEC 2015 test suite. Then the performance of CLJAYA is investigated by extracting the unknown parameters of three PV models including single diode model, double diode model and PV module model. Experimental results prove the superiority of CLJAYA on these test cases in terms of accuracy and efficiency by comparing with Jaya algorithm and other competitive algorithms.