Novel parameter identification for complex solar photovoltaic models via dynamic L-SHADE with parameter decomposition

The determination of unknown parameters of photovoltaic models has always been a difficult problem in the field of nonlinear optimization, which is inseparable from the energy conversion efficiency of photovoltaic power generation systems. Due to the high degree of nonlinearity and difficulty in parameter identification of photovoltaic models with multiple diode branches, this paper proposes a novel and efficient evolutionary algorithm based on success-history adaptation differential evolution with linear population size reduction (L-SHADE)—Dynamic L-SHADE with parameter decomposition method (DaL-SHADE) to identify unknown parameters of photovoltaic models. In DaL-SHADE, a dynamic crossover rate sorting technology is introduced, which establishes the relationship between the population and its crossover rate that is ignored in L-SHADE, and thereby the optimization ability of the algorithm to identify unknown parameters is improved. Secondly, the unknown parameters of different types of photovoltaic models are decomposed into linear parameters and nonlinear parameters, so that the dimensionality of the photovoltaic model is reduced. Therefore, DaL-SHADE not only has efficient optimization capabilities but also reduces the dimensionality of different types of photovoltaic models. Through the examination of four photovoltaic models with different degrees of nonlinearization, root mean square error obtained by DaL-SHADE is better than that of the comparison algorithms.