基于多尺度特征提取的IES多元负荷短期联合预测  

Multi-load Short-term Joint Forecasting of Integrated Energy System Based on Multi-scale Feature Extraction

作　　者：沈赋 刘思蕊 徐潇源 王健 单节杉[1] 翟苏巍 SHEN Fu;LIU Sirui;XU Xiaoyuan;WANG Jian;SHAN Jieshan;ZHAI Suwei(Faculty of Electric Power Engineering,Kunming University of Science and Technology,Kunming 650500,China;School of Electric Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Electric Power Research Institute,Yunnan Power Grid Co.,Ltd.,Kunming 650521,China)

机构地区：[1]昆明理工大学电力工程学院,昆明650500 [2]上海交通大学电子信息与电气工程学院,上海200240 [3]云南电网有限责任公司电力科学研究院,昆明650521

出　　处：《高电压技术》2024年第7期2918-2930,共13页High Voltage Engineering

基　　金：国家自然科学基金(52107097);云南省兴滇英才支持计划(KKRD202204021);云南省应用基础研究计划(202101BE070001-061,202201AU070111);昆明理工大学高层次人才平台建设项目(KKZ7202004004)。

摘　　要：为提高综合能源系统(integrated energy system,IES)多元负荷预测的精确度,综合考虑多能源相互作用机理、多元负荷耦合特性及气象因素相关性,提出了一种基于多尺度特征提取的IES多元负荷短期联合预测方法。首先,通过最大互信息系数(maximum information coefficient,MIC)研究多元负荷耦合特性及影响因素相关性,选择预测特征;其次,利用变分模态分解技术(variational mode decomposition,VMD)对输入特征进行分解,提升特征纯洁度;最后,采用卷积神经网络-双向长短期记忆神经网络(convolutional neural network-bidirectional long and short-term memory,CNN-BiLSTM)多任务学习模型进行纵向、横向特征选择,注意力(Attention)机制对重要特征差异化提取,实现多尺度特征提取,并利用雪消融优化器(snow ablation optmizer,SAO)对VMD和CNN-BiLSTM多任务学习模型进行超参数优化,以此实现IES多元负荷的联合预测。以美国亚利桑那州实测数据进行实验,结果表明,无论与单一预测方法还是与其他模型相比,所提联合预测方法的均方根误差更低、准确率更高,在IES多元负荷预测中具有更高的精确性和鲁棒性。A multi-load short-term joint forecasting model of integrated energy system is proposed to improve the accu-racy of multi-load forecasting of integrated energy system based on multi-scale feature extraction by comprehensively considering the interaction mechanism of multi-energy,the coupling characteristics of multi-load and the correlation of meteorological factors.Firstly,the coupling characteristics of multi-load and the correlation of influencing factors are studied by the maximum information coefficient,and the prediction characteristics are selected.Secondly,the input fea-tures are decomposed by variational modal decomposition technology to enhance features purity.Finally,the CNN-BiLSTM multi-task learning model is used for feature fusion,and the Attention mechanism is used to select im-portant features differently to realize multi-scale feature extraction.In addition,hyperparameter optimization of the VMD and CNN-BiLSTM multi-task learning model is achieved by the snow ablation optimizer to realize joint forecasting of IES multivariate loads.Experiments were conducted using real-world data from Arizona,USA.The results indicate that the proposed joint forecasting method possesses lower root mean square error and higher accuracy compared with single forecasting method or other models and greater robustness in IES multivariate load forecasting.

关 键 词：联合预测 多尺度特征提取 综合能源系统 多元负荷 多任务学习 雪消融优化器 

分 类 号：TM715[电气工程—电力系统及自动化]