Hybrid Symbiotic Differential Evolution Moth-Flame Optimization Algorithm for Estimating Parameters of Photovoltaic Models

Abstract

Obtaining suitable parameters of photovoltaic models based on measured current-voltage data of the PV system is vital for assessing, controlling, and optimizing photovoltaic systems. To acquire specific parameters of photovoltaic models, we proposed a meta-heuristic algorithm named hybrid symbiotic differential evolution moth-flame optimization (HSDE-MFO) algorithm. The proposed algorithm implements our new proposed symbiotic algorithm structure (SAS). This structure is inspired by soybean-rhizobium nodule symbiosis in nature. The proposed SAS divides the population into two parallel working subgroups , i.e., soybean group and rhizobium group. Soybean group that focuses on exploration is updated by the strategies in the DE algorithm; the rhizobium group that emphasizes on exploitation is renewed by the strategies in the MFO algorithm. Artificial particle selection strategy and artificial flames generation strategy are developed to generate high-quality mutant materials and high-quality flames, respectively. The above-proposed methods balance the exploration ability and exploitation ability and ensure a bionic structure of the proposed algorithm. Moreover, a new elite strategy is developed to offer a chaotic particle to further refine the quality of the current population. The proposed HSDE-MFO is employed to solve the parameters identification problem of photovoltaic models, i.e., single diode, double diode, and photovoltaic module and compared with recently wellestablished algorithms. Experimental results indicate that HSDE-MFO can acquire precise parameters of the three photovoltaic models and stable performance in 30 independent runs. INDEX TERMS Photovoltaic (PV)，moth flame optimization algorithm (MFO), differential evolution algorithm (DE), Parameter identification, Soybean-rhizobium nodule symbiosis.I. INTRODUCTION

Moth Flame Optimization: Theory, Modifications, Hybridizations, and Applications

“…Wu et al [ 63 ] introduced a new population-based algorithm by integrating hybrid symbiotic DE and MFO to acquire suitable PV model parameters. The proposed method is known as the HSDE-MFO algorithm and has three stages.…”

Solution of Mixed-Integer Optimization Problems in Bioinformatics with Differential Evolution Method

“…Hybrid algorithms were introduced with intent to integrate the advantages of DE and other meta-heuristic algorithms, such as a hybrid DE with the gaining-sharing knowledge algorithm (GSK) and Harris hawks optimization (HHO), DEGH [75], a hybrid artificial bee colony with DE (HABCDE) [76], a semi-parametric adaptation method in the LSHADE hybridized with covariance matrix adaptation evolution strategy (LSHADE-SPACMA) [77], a hybrid algorithm based on self-adaptive gravitational search algorithm (SGSADE) [78], a hybrid algorithm for DE and particle swarm optimization (DEPSO) [79], mixed DE with whale optimization algorithm (MDE-WOA) [80], a hybrid adaptive teaching-learningbased optimization algorithm with DE (ATLDE) [81], a hybrid symbiotic DE moth flame optimization algorithm (HSDE-MFO) [82], a modified Boltzmann annealing [83] differential evolution algorithm (BADE) [84], an adaptive DE with PSO (A-DEPSO) [85], a hybrid differential symbiotic organism search (HDSOS) algorithm [86], a new local search scheme based on the Hadamard matrix (HLS) [87], an opposition-based learning DE (ODE) [88], DE/EDA [89], which combined DE with the estimation of distribution algorithm, a DE variant with commensal learning and uniform local search, named CUDE [90], and ESADE [91], which combines simulated annealing in the selection stage.…”

Improved Complete Ensemble Robust Local Mean Decomposition With Adaptive Noise for Slewing Bearings Performance Degradation Assessment

“…However, the randomness of hyper parameter (noise amplitude and ensemble trials) is a major drawback of noise-assisted approach [24]. Moth-flame optimization (MFO), which is a relatively new intelligent optimization algorithm [25], has been widely applied in many field [26], [27]. Therefore, the proposed method named improved MFO-CERLMDAN exploits the merits of CERLMDAN approach to handle multiscale signals and the benefits of MFO method to handle parameter optimization.…”