基于相似日和PSO-LightGBM的超短期电力负荷预测方法

Ultra-Short-Term Power Load Forecasting Method Based on Similar Days and PSO-LightGBM

为提高超短期电力负荷预测的准确度,提出一种基于相似日和粒子群算法——轻量梯度提升机的超短期电力负荷预测方法。对初始数据集进行特征构建,并利用灰色关联法筛选出与待预测日关联度较高的历史日。使用粒子群优化算法对LightGBM算法进行参数寻优,构造PSO-LightGBM负荷预测模型。实验分析表明,相较于传统预测方法,该方法提高了电力负荷预测的精度,为电力系统稳定运行提供了保障。

In order to improve the accuracy of ultra-short-term power load forecasting,an ultra-short-term power load forecasting method based on similar days and particle swarm optimization-lightweight gradient hoist is proposed.Construct features of the initial data set,and use the grey correlation method to filter out the historical days with a higher degree of correlation with the days to be predicted.The particle swarm optimization algorithm is used to optimize the parameters of the LightGBM algorithm,and the PSO-LightGBM load forecasting model is constructed.Experimental analysis shows that compared with traditional forecasting methods,this method improves the accuracy of power load forecasting and provides a guarantee for the stable operation of the power system.