【摘要】 针对Rao-1算法在光伏模型参数辨识中存在求解精度低、收敛速度慢、易陷入局部最优等问题,提出了一种基于广义对立学习(generalized opposition-based learning, GOBL)的改进Rao-1算法(GORao-1算法)。首先,采用2种不同的更新策略,种群个体根据适应度函数值排名选择相对应的更新公式,充分利用不同类型个体的信息,提高了算法的收敛速度和全局搜索能力。其次,引入GOBL策略,避免了算法陷入局部最优。最后,将GORao-1算法应用到Photo Watt-PWP201光伏组件模型的参数辨识中,并将辨识结果与其他7种优化算法的辨识结果进行对比,对比结果表明该算法在准确性和收敛速度方面优于其他算法;利用不同辐照度和不同温度下S75多晶硅光伏组件的实测数据进行实验,实验结果进一步验证了该算法能在不同环境下准确有效地进行参数辨识。更多还原

【Abstract】 In view of the problems of Rao-1 algorithm in parameter identification of photovoltaic model, such as low solution accuracy, slow convergence speed and easy to fall into local optimum, an improved Rao-1 algorithm(GORao-1 algorithm) based on generalized opposition-based learning(GOBL) is proposed. Firstly, two different updating strategies are adopted, and population individuals select the corresponding updating formula according to the ranking of fitness function values, which makes full use of the information of different types of individuals,so that the convergence speed and global search ability of the algorithm are greatly improved. Then, the GOBL strategy is introduced to avoid the algorithm falling into local optimum. Finally, the GORao-1 algorithm is applied to the parameter identification of Photo Watt-PWP 201 photovoltaic module model, and the identification results are compared with those of other 7 optimization algorithms. The comparison results show that the proposed algorithm is superior to the other algorithms in terms of accuracy and convergence speed. The measured data of S75 polysilicon photovoltaic modules under different irradiance and temperature are used for experiments. The experimental results further verify that the proposed algorithm can accurately and effectively identify parameters under different environments.更多还原