Research On Parameter Identification Of Solar Photovoltaic System Based On Improved Swarm Intelligence Optimization

It is very important for the simulation,evaluation,control and optimization of photovoltaic system to build high-precision models for solar cells and photovoltaic modules based on experimental data.Thus,efficient algorithms are needed to obtain the best parameters reliably to establish the optimal model.The population-based optimization algorithm is currently considered to be a more promising technology.In this paper,Harris hawk optimization algorithm,Sine cosine algorithm and Whale optimization algorithm are improved,and these three improved algorithms are used to identify parameters of solar cells and photovoltaic modules.Firstly,Harris hawk optimization algorithm is a novel optimizer.Aiming at making up for the shortcomings of the algorithm in the parameter identification of photovoltaic system,this paper introduces the orthogonal learning mechanism and general opposition-based learning strategy to the original algorithm.In the orthogonal learning mechanism,based on the idea of orthogonal experiment design with the quantization orthogonal crossover operator as well as extremum difference analysis,the local search ability of the algorithm is strengthened while the solution quality is improved.On this basis,the introduction of the general opposition-based learning strategy enhances the diversity of the population and improves the exploration ability of the algorithm,which also can effectively avoid the algorithm falling into a stagnation state prematurely.Secondly,a new algorithm is proposed by combining the since cosine algorithm and the development strategy based on Nelder-Mead simplex method and the diversification mechanism which is named oppositionbased learning.When using the core strategy of the original cosine optimization algorithm,the simplex method is used to improve the accuracy of the algorithm and enhance the local search ability of the algorithm.At the same time,the opposition-based learning is used to explore the search space and improve the global search ability of the algorithm.As a result,the proposed new method achieves a more stable balance between exploitation and exploratory trends.Finally,a new improved approach is proposed for whale optimization algorithm.The improved whale optimization algorithm adds Levy flight mechanism and pattern search operator on the basis of the original algorithm.In the new algorithm proposed in this paper,the population renewal formula of the original algorithm is used to find the current optimal solution of the problem,and Levy flight is used to maintain the diversity of solutions.Thus,the exploration of the algorithm is enhanced.Then,the pattern search mechanism is utilized to exploit the neighborhood of the current optimal solution in search of a better one.The combination of the two methods not only reduces the possibility of the algorithm falling prematurely into the local optimum,but also improves the quality of solution.Of course,the three improved algorithms proposed in this paper are applied to identify the uncertain parameters of solar cell single diode model,double diode model and photovoltaic module.The experimental results clearly show the effectiveness and accuracy of the three algorithms.In addition,for testing in different types of photovoltaic modules,it demonstrates that the proposed algorithm can give excellent results at different temperature and different light intensity.Therefore,the three improved algorithms proposed in this paper can be used as a new choice for parameter identification of solar cells and photovoltaic modules.