基于WHO-AGRU-MTL的综合能源系统多元负荷预测  

Multi Factor Load Forecasting of Integrated Energy System Based on WHO-AGRU-MTL

作　　者：曲坤宁 QU Kunning(Ningxia Jiaotou Technology Development Co.,Ltd.,Yinchuan 7500,China)

机构地区：[1]宁夏交投科技发展有限公司,宁夏银川750000

出　　处：《山东冶金》2024年第5期47-51,共5页Shandong Metallurgy

摘　　要：为了进一步提高IES短期负荷预测的准确性,提出了基于野马优化算法的短期负荷预测模型(WHO-AGRU)用于IES的电、冷、热负荷。其中,基于注意力机制的门控循环单元(AGRU)可以从多个时间维度提取IES负荷数据中隐含的耦合和周期特性。与其他预测模型相比,算例结果表明,在电力、供暖和制冷的短期多元负荷预测性能方面,本研究所提出的模型具有更高的预测精度,RMSE平均降低69.93%,MAPE降低了62.10%,模型耗费时间平均缩短了37.95%。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 

分 类 号：TK01[动力工程及工程热物理] TM715[电气工程—电力系统及自动化]