基于多变量灰色遗传模型的配电网柔性中长期负荷曲线化预测方法

Flexible Medium and Long-term Load Curve Prediction Method for Distribution Network Based on Multivariable Grey Genetic Model

传统配电网柔性中长期负荷曲线化预测方法,缺乏对配电网负荷预测权重的合理有效分配,导致预测准确性不高。因此,提出基于多变量灰色遗传模型的配电网柔性中长期负荷曲线化预测方法。使用随时间反向传播(BPTT)算法训练处理后的配电网负荷数据,得到配电网负荷特征。在考虑负荷数据特征基础上,使用多变量灰色遗传模型分配负荷曲线预测权重。根据预测权重,聚类处理负荷增长等参数,输出配电网柔性中长期负荷预测曲线,从而曲线化预测配电网柔性中长期负荷。实验结果表明,设计的方法对电网柔性中长期负荷的预测误差较小、准确性较高,满足配电网柔性中长期负荷曲线化预测的需求。

The traditional flexible medium to long term load curve prediction method for distribution networks lacks a reasonable and effective allocation of load prediction weights,resulting in low prediction accuracy.Therefore,a multivariable grey genetic model based flexible medium and long-term load curve prediction method for distribution networks is proposed.Train the processed distribution network load data using the back-propagation through time(BPTT)algorithm to obtain the distribution network load characteristics.On the basis of considering the characteristics of load data,a multivariate grey genetic model is used to allocate weights for load curve prediction.According to the predicted weights,cluster processing parameters such as load growth,output the flexible medium and long-term load prediction curve of the distribution network,and thus curvaturally predict the flexible medium and long-term load of the distribution network.The experimental results show that the designed method has a smaller prediction error and higher accuracy for the medium to long-term flexible load of the power grid,meeting the demand for curved prediction of the flexible medium and long-term load of the distribution network.