针对局部阴影状态下光伏阵列P-U特性曲线多峰值的问题，提出了一种基于搜索-计算-判断的三步法全局最大功率点跟踪(GMPPT)策略。该方法使用改进型扰动观察法反向搜索第一个峰值点，利用计算和判断减小搜索区域，将多峰值间接转化为单峰值搜索，解决了搜索过程中陷入局部极值的问题。通过建立3种阴影状态下的模型，将该方法与现有最大功率点跟踪(MPPT)算法进行仿真对比，并建立相应的试验进行验证。仿真及试验结果表明，该方法在有、无阴影状态下均能稳定实现GMPPT，有效地提高了系统的跟踪精度和跟踪速度。

PV array P-U characteristic curve in partial shadow state has multiple peaks. To solve this problem, a three-step global maximum power point tracking (GMPPT) strategy based on search-calculation-determination is proposed. The improved disturbance observation method is used to search the first peak point in reverse direction, and the search area is reduced by calculation and judgment. The multi-peak value search can be indirectly transformed into a single peak value search, which overcomes the problem of falling into local extremum in the search process. Through the establishment of three kinds of shadow state models, this method is compared with the existing maximum power point tracking (MPPT) algorithm, and the corresponding experiments are established. The simulation and experimental results show that the method can stably realize the GMPPT in the state of shadow and noshadow, and effectively improve the tracking accuracy and speed.

上海市闵行区科技项目(2018MH303)