Research On Intelligent Optimization Algorithm For Photovoltaic Model Parameter Identification

Photovoltaic systems(PV)can convert solar radiation into electrical energy,making them a popular choice for lighting and heating due to their simple structure and ease of installation and maintenance.PV models can effectively describe the nonlinear behavior of these systems,and their unknown parameters influence the accuracy of these models.These parameters are critical not only for performance evaluation and quality control of PV models,but also for PV systems’ maximum power point tracking.As such,accurate and efficient identification of the unknown parameters of PV models is crucial.Due to PV models’ multimodal and nonlinear nature,traditional analytical methods are prone to falling into local optima and have low search accuracy when identifying parameters.Therefore,this paper proposes two improved intelligent optimization algorithms for identifying the parameters of PV models.The research in this paper consists of the following two parts:Dual-feedback adaptive clone selection algorithm with golden sinusoidal search(DCSAGS)is designed to address conflicts between clone size and convergence speed in clone selection algorithms.In this algorithm,a dual feedback adaptive cloning strategy is proposed to control the adaptive change of cloning factor by two indicators,namely,the difference in population fitness and the population harmonic mean distance,to dynamically adjust the clone size and solve the conflict between clone size and convergence speed.In addition,an elite learning strategy based on the golden sine is proposed further to improve the accuracy of the clone selection algorithm to extract the unknown parameters of PV components.In this paper,the proposed algorithm is used to identify the parameters of two equivalent models of monocrystalline silicon and multi-crystalline silicon PV modules on single and double diode module models under fixed conditions,and the experiments are compared with nine advanced algorithms,which demonstrate that the proposed algorithm can accurately and reliably solve the parameter identification problem of different PV module models under fixed environmental conditions.An improved artificial ecosystem optimization(IAEO)algorithm that considers light and temperature variations is proposed to address the shortcomings of most existing parameter identification methods that do not consider the effect of environmental changes on parameter identification performance.In the proposed algorithm,a nonlinear control parameter adjustment strategy is proposed to replace the linear weight coefficients in the original artificial ecosystem with nonlinear control parameters to better balance the relationship between exploration and exploitation by exploiting the ergodic and non-repetitive nature of chaos.To verify the performance of the proposed algorithm,the proposed algorithm is applied to the parameter identification of PV models under fixed environmental conditions(RTC France mono-crystalline PV cells and Photowat PWP201 multi-crystalline PV modules)and non-fixed environmental conditions(commercial thin-film ST40,mono-crystalline SM55 and multicrystalline KC200 GT PV modules),a and compare the experiments with six advanced algorithms.The results demonstrate that the proposed algorithm can be an effective solution to the problem of parameter identification of multiple PV cell models under fixed and non-fixed environmental conditions.