【摘要】 为提升光伏功率预测的准确度和可靠性，本文提出一种基于改进灰狼优化算法（LGWO）、变分模态分解法（VMD）、粒子群优化算法（PSO）、长短期记忆网络（LSTM）组成的短期光伏辐照度预测模型。首先，通过LGWO对VMD关键分解参数K和α₁进行优化构建参数优化模型，然后利用LSTM对VMD优化后分解的各模态参数进行预测。同时利用改进的PSO以均方根误差为目标函数，优化LSTM隐藏层数量、LSTM迭代次数、LSTM学习率等神经网络参数，最终实现模型预测准确性与可靠性的提升。最后，对甘肃地区某光伏场站出力建模并进行预测分析，结果表明对比传统的LSTM预测模型，本文算法对预测的准确度有较大提升。更多还原

【Abstract】 To enhance the accuracy and reliability of photovoltaic power prediction, this paper proposes a short-term photovoltaic irradiance prediction model based on the improved grey wolf optimizer（GWO）, variational mode decomposition（VMD）, particle swarm optimization（PSO）, and long short term memory（LSTM）. First, LGWO optimizes the key decomposition parameters K and α₁ of VMD to construct a parameter optimization model. Then, LSTM predicts each mode parameter decomposed by the optimized VMD. Simultaneously, the improved PSO, using root mean square error as the objective function, optimizes the neural network parameters, including the number of LSTM hidden layers, LSTM iterations, and LSTM learning rate. This ultimately improves the prediction accuracy and reliability of the model. Finally, the output modeling and prediction analysis of a photovoltaic power plant in Gansu Province are conducted. The results show that, compared with the traditional LSTM prediction model, the proposed algorithm significantly improves the prediction accuracy.更多还原