Photovoltaic power prediction method and system based on EEMD-FE-IWOA-BiLSTM

The invention discloses a photovoltaic power prediction method and system based on EEMD-FE-IWOA-BiLSTM. The method comprises the following steps: introducing a Pearson correlation coefficient to analyze the seasonal distribution characteristics of important influence factors of photovoltaic power generation; data set characteristic decomposition based on ensemble empirical mode decomposition and fuzzy entropy; the WOA is improved from three aspects of population initialization, a position updating strategy and prevention of falling into local optimum; an improved IWOA algorithm is adopted to optimize the BiLSTM; and carrying out photovoltaic power prediction by using the optimized IWOA-BiLSTM hybrid model. The method is high in prediction precision and good in implementation and application effects, and prediction results meet actual application requirements. 本发明公开了一种基于EEMD-FE-IWOA-BiLSTM的光伏功率预测方法及系统，引入Pearson相关系数分析光伏发电重要影响因子的季节分布特征；基于集合经验模态分解及模糊熵的数据集特征分解；从种群初始化、位置更新策略以及预防陷入局部最优这三个方面对WOA进行改进；采用改进的IWOA算法对BiLST