阐述了负荷预测的应用研究,包含光伏电池模块、MPPT模块、BOOST模块、逆变模块，粒子图像分割及匹配均为自行编制的子例程，利用自然梯度算法，迭代自组织数据分析，结合PCA的尺度不变特征变换(SIFT)算法。- It describes the application of load forecasting, PV modules contain, MPPT module, BOOST module, inverter module, Particle image segmentation and matching subroutines themselves are prepared, Use of natural gradient algorithm, Iterative self-organizing data analysis, Combined with PCA scale invariant feature transform (SIFT) algorithm.