基于SVR与扰动观察法的光伏阵列多峰值MPPT研究

摘要

当光伏组件出现局部阴影遮挡或光照不均匀时,光伏阵列的输出特性将发生改变,此时的P-U特性曲线将呈现多峰值现象,传统的基于单峰P-U特性曲线的MPPT算法将失效,很难准确地跟踪到全局的最大功率点.为解决该问题,提出了一种基于支持向量机回归与扰动观察法的MPPT融合算法.利用支持向量机的全局优化、泛化性能高的特点,结合扰动观察法的控制简单、容易实现的优点来实现最大功率点的跟踪.仿真结果表明,在真实的光照、温度及光照突变等外界条件下,该新型融合算法与传统的扰动观察法相比,光伏阵列在局部阴影下不会陷于局部峰值,能迅速准确地搜寻到全局最大功率点.%When PV module is under local shadow or non-uniform illumination conditions,the output characteristics of PV arrays will change.P-U characteristic curve presents multiple peaks;traditional MPPT algorithm based on single peak P-U curve may fail;it is difficult to accurately track the global maximum power point.To address this drawback,a MPPT fusion algorithm based on combination of perturbation method and SVM was put forward.The global optimization and high generalization performance characteristics of support vector machine combining with easy control and realization of perturbation method were used to achieve maximum power point tracking.The simulation results show that in real external conditions,such as light,temperature and illumination mutation,compared with traditional perturbation method,the new fusion algorithm can not be trapped in local peak,and the system can quickly and accurately track the global maximum power point.