基于VMD和PSO-SVR的短期电力负荷多阶段优化预测

为降低短期负荷序列的非线性以提升预测精度，提出一种基于多阶段优化的变分模态分解（variational mode decomposition, VMD）和粒子群算法优化支持向量回归（particle swarm optimization support vector regression, PSO-SVR）的短期电力负荷预测模型。第1阶段采用VMD优化和预处理原始负荷序列，分解获得多个较为平稳的模态分量。第2阶段利用相空间重构优化重组各序列分量，并针对各分量分别建立支持向量回归（support vector regression，SVR）预测模型。第3阶段将粒子群算法（particle swarm optimization，PSO）用于优化SVR模型内部参数，便于更好地进行训练和预测。最后累加所有序列的预测值，实现短期电力负荷预测。研究结果表明：所提方法可以取得更高的预测精度。

Multi-stage Optimization Forecast of Short-term Power Load Based on VMD and PSO-SVR

To reduce the non-linearity of the short-term load sequence and improve the prediction accuracy, a short-term load forecasting model based on multi-stage optimization variational mode decomposition (VMD) and particle swarm optimization optimize support vector regression (PSO-SVR) is proposed. In the first stage, VMD optimization and pre-processing of the original load sequence are used to decompose and obtain multiple relatively stable modal components. In the second stage, phase space reconstruction is used to optimize and reorganize each sequence component, and establish support vector regression（SVR）prediction model for each component. In the third stage, the particle swarm optimization（PSO）algorithm is applied to optimize the internal parameters of the SVR model to facilitate better training and forecasting. Finally, the predicted values of all sequences are accumulated to realize the short-term power load forecast. The results show that the proposed method can achieve higher prediction accuracy.