自适应BP神经网络在光伏MPPT中的应用

Self-adaption BP neural network application in photovoltaic maximum-power-point tracking

首先介绍了光伏电池的特性,并在比较了扰动观察法、电导增量法、恒定电压法、开路电压法等几种光伏系统最大功率跟踪(MPPT)算法的基础上,提出了一种基于BP神经网络的最大功率跟踪的控制策略。该策略将温度和光强作为输入变量,通过神经网络识别后可得到最大功率点。仿真结果表明,该方法能够快速、准确地跟踪光伏电池的最大功率点,具有良好的控制精度和适应性,显著提高了光伏系统的转换效率。

The characteristics of photovoltaic cells were described. A control scheme based on BP neural network was proposed for MPPT through comparing several MPPT methods, including perturbation and observation method （P＆O）, incremental conductance method, constant voltage method, open circuit voltage and so on. Taking temperature and light intensity as the input variables in this strategy, after identification of the neural network, the maximum power point could be obtained. Simulation results show that this method can quickly and accurately track the maximum power point of photovoltaic cells with good control accuracy and adaptability, significantly improving the conversion efficiency of photovottaic system.