An enhanced adaptive comprehensive learning hybrid algorithm of Rao-1 and JAYA algorithm for parameter extraction of photovoltaic models

Mathematical Biosciences and Engineering

Yufei Wang¹  Juan Zhao¹  Yujun Zhang¹  Shuijia Li²  Zhengming Gao³  Fengjuan Yao⁴  Liuwei Tao⁵  Yuxin Yan⁶ 

[1] 1. School of electronics and information engineering, Jingchu University of Technology, Jingmen 448000, China;2. School of Computer Science, China University of Geosciences, Wuhan 430074, China;3. School of computer engineering, Jingchu University of Technology, Jingmen, 448000, China6. Institute of intelligent information technology, Hubei Jingmen industrial technology research institute, Jingmen 448000, China;3. School of computer engineering, Jingchu University of Technology, Jingmen, 448000, China;4. School of foreign languages, Jingchu University of Technology, Jingmen, 448000, China;5. Academy of arts, Jingchu University of Technology, Jingmen 448000, China;

关键词: parameter extraction;    photovoltaic models;    jaya algorithm;    rao-1 algorithm;    ehrjaya algorithm;   

DOI  :  10.3934/mbe.2022263

来源: DOAJ

【 摘 要 】

In order to maximize the acquisition of photovoltaic energy when applying photovoltaic systems, the efficiency of photovoltaic system depends on the accuracy of unknown parameters in photovoltaic models. Therefore, it becomes a challenge to extract the unknown parameters in the photovoltaic model. It is well known that the equations of photovoltaic models are nonlinear, and it is very difficult for traditional methods to accurately extract its unknown parameters such as analytical extraction method and key points method. Therefore, with the aim of extracting the parameters of the photovoltaic model more efficiently and accurately, an enhanced hybrid JAYA and Rao-1 algorithm, called EHRJAYA, is proposed in this paper. The evolution strategies of the two algorithms are initially mixed to improve the population diversity and an improved comprehensive learning strategy is proposed. Individuals with different fitness are given different selection probabilities, which are used to select different update formulas to avoid insufficient using of information from the best individual and overusing of information from the worst individual. Therefore, the information of different types of individuals is utilized to the greatest extent. In the improved update strategy, there are two different adaptive coefficient strategies to change the priority of information. Finally, the combination of the linear population reduction strategy and the dynamic lens opposition-based learning strategy, the convergence speed of the algorithm and ability to escape from local optimum can be improved. The results of various experiments prove that the proposed EHRJAYA has superior performance and rank in the leading position among the famous algorithms.

【 授权许可】

Unknown