基于改进SCSO算法的光伏MPPT研究

Research on PV MPPT based on improved SCSO algorithm

在解决光伏阵列在局部遮挡时发电效率降低的问题时,传统最大功率点追踪(MPPT)方法容易追踪失败。为此,提出一种改进沙猫群优化算法的最大功率点追踪方法。该算法在标准沙猫群算法的基础上,引入了精英反向学习和自适应t分布,同时优化沙猫群算法(SCSO)的局部搜索并融合Jaya算法。通过对4种典型单峰、多峰函数的测试,证明该算法具有极高的收敛速度,容易跳出局部最优值。将算法应用于MPPT控制中,仿真结果表明:在静态遮荫情况下,所提方法的搜索最大功率点的时间更少;在动态遮荫条件下,重新搜寻到最大功率点的响应时间平均为0.2 s。实验表明所提算法可以适应动态变化的天气,解决了传统算法收敛速度和防止陷入局部最优等问题。

In allusion to the problem that the power generation efficiency of photovoltaic arrays can decrease under local occlusion,and traditional maximum power point tracking(MPPT)are prone to tracking failure,a MPPT method based on improved sand cat swarm optimization(SCSO)algorithm is proposed.In this algorithm,the elite backward learning and adaptive t-distribution are intorduced on the basis of the standard sand cat swarm algorithm,the local search is optimized,and the Jaya algorithm is intergated.By testing four typical single-peak and multi-peak functions,it is proved that the algorithm has faster convergence speed and is prone to jumping out of local optima.The algorithm is applied into MPPT control,and the simulation results show that under static shading,the proposed method has less time to search for the maximum power point;under dynamic shading conditions,the average response time to rediscover the maximum power point is 0.2 s.The experiments show that the proposed algorithm can adapt to dynamically changing weather,can improve the convergence speed of traditional algorithms,and can prevent getting stuck in local optima.