In order to further improve the accuracy of IES short-term load forecasting,a Integrated Energy System short-term load forecasting model based on the Wild Horse Optimization Algorithm(WHO-AGRU)was proposed for IES'electricity,cooling,and heating loads.Among them,the attention based Gated Recurrent Unit(AGRU)can extract the coupling and periodic characteristics hidden in IES load data from multiple time dimensions.Compared with other prediction models,the calculation results show that the model proposed in this study has higher prediction accuracy in short-term multi load forecasting performance of electricity,heating,and cooling.The average RMSE is reduced by 69.93％,MAPE is reduced by 62.10％,and the average model time is shortened by 37.95％.
关键词
综合能源系统/多元负荷预测/野马优化算法/多任务学习算法/预测精度
Key words
integrated energy system/multivariate load forecasting/WHO-AGRU/MTL/prediction accuracy
维普
万方数据