陈东海,马旭,王波,朱晓杰,白文博
CHEN Donghai,MA Xu,WANG Bo,ZHU Xiaojie,BAI Wenbo

• 1:国网浙江省电力有限公司宁波供电公司
• 2:国网浙江余姚市供电有限公司
• 3:宁波市电力设计院有限公司

摘要(Abstract)：

建筑空调负荷预测对于提升建筑用电负荷和区域配电网电力负荷预测的准确性具有重要意义。为了提高建筑空调负荷预测的准确性，提出一种基于多阶段注意力机制的建筑空调负荷预测方法。首先，构建影响因素注意力模块，以充分考虑不同影响因素对于建筑空调负荷预测的重要性差异；然后，采用长短期记忆神经网络模型提取各个时刻影响因素的隐含特征；最后，构建时间注意力模块，将不同时刻的影响因素隐含特征根据建筑空调负荷预测的重要性加以区分，以求得最终的空调负荷预测结果。算例结果表明，影响因素注意力模块和时间注意力模块的构建，都有利于提高模型对建筑空调负荷的拟合能力，从而有效提升建筑空调负荷预测的准确性。
Load forecasting of building air conditioning is of great significance for improving the accuracy of power load forecasting of buildings and regional distribution networks. In order to improve the load forecasting accuracy of building air conditioning, a load forecasting method based on multi-stage attention mechanism is proposed. First, the attention module of influencing factors is constructed to fully consider the importance difference of different influencing factors for load forecasting of building air conditioning. Second, the LSTM network model is used to extract the implicit features of influencing factors in each hour. Finally, the temporal attention module is constructed to differentiate the implicit features of the influencing factors in different hours according to the importance of building air conditioning load forecasting to obtain the results of the air conditioning load forecasting. The example results show that the construction of the influencing factor attention module and the temporal attention module are both conducive to improving the model's ability to fit the building air-conditioning load, thus effectively improving the load prediction accuracy of building air-conditioning.

关键词(KeyWords)： 建筑空调负荷;多阶段注意力机制;长短期记忆神经网络;负荷预测
building air conditioning load;multi-stage attention mechanism;LSTM network;load forecasting

Abstract:

Keywords:

基金项目(Foundation): 宁波永耀电力投资集团有限公司科技项目（CY820400QT20210652）

作者(Author): 陈东海,马旭,王波,朱晓杰,白文博
CHEN Donghai,MA Xu,WANG Bo,ZHU Xiaojie,BAI Wenbo

DOI: 10.19585/j.zjdl.202310007

参考文献(References)：

• [1]陈健，张维桐，张逸成，等.考虑不同空调负荷特性的微网双层优化调度[J].电网技术，2018,42(5):1424-1431.CHEN Jian,ZHANG Weitong,ZHANG Yicheng,et al.Bi-level optimal dispatch of microgrid considering different air-conditioning load characteristics[J]. Power System Technology,2018,42(5):1424-1431.
• [2]李泽宁，苏粟，靳小龙，等.面向提高风电接纳能力的智慧建筑能量管理策略[J].电网技术，2021,45(6):2288-2298.LI Zening,SU Su,JIN Xiaolong,et al. Energy management strategy of smart buildings for improving wind power accommodation ability[J]. Power System Technology,2021,45(6):2288-2298.
• [3]刘开欣，王丹，贾宏杰，等.基于参数序列化的中央空调变负载率调节控制策略研究[J].电网技术，2017,41(9):3041-3050.LIU Kaixin,WANG Dan,JIA Hongjie,et al.Research on an variable load rate control strategy for central airconditioner based on parameter serialization[J]. Power System Technology,2017,41(9):3041-3050.
• [4]庄家懿，杨国华，郑豪丰，等.并行多模型融合的混合神经网络超短期负荷预测[J].电力建设，2020,41(10):1-8.ZHUANG Jiayi,YANG Guohua,ZHENG Haofeng,et al.Ultra-short-term load forecasting using hybrid neural network based on parallel multi-model combination[J].Electric Power Construction,2020,41(10):1-8.
• [5]刘伟，张锐锋，彭道刚.基于K-Adaboost数据挖掘的配电网负荷预测[J].浙江电力，2019,38(1):104-110.LIU Wei,ZHANG Ruifeng,PENG Daogang.Load forecasting of distribution network based on K-adaboost data mining[J].Zhejiang Electric Power,2019,38(1):104-110.
• [6]周璇，凡祖兵，刘国强，等.基于多元非线性回归法的商场空调负荷预测[J].暖通空调，2018,48(3):120-125.ZHOU Xuan,FAN Zubing,LIU Guoqiang,et al.Cooling load prediction of shopping mall air conditioning based on multivariate nonlinear regression method[J].Heating Ventilating&Air Conditioning,2018,48(3):120-125.
• [7]王威，王波，何宇，等.基于负荷分解和灰色理论的电网最大负荷预测[J].浙江电力，2018,37(8):48-53.WANG Wei,WANG Bo,HE Yu,et al. Maximum grid load forecasting based on load decomposition and grey theory[J].Zhejiang Electric Power,2018,37(8):48-53.
• [8]刘文杰，刘禾，王英男，等.基于完整自适应噪声集成经验模态分解的LSTM-Attention网络短期电力负荷预测方法[J].电力建设，2022,43(2):98-108.LIU Wenjie,LIU He,WANG Yingnan,et al.Short-term power load forecasting method based on CEEMDAN and LSTM-attention network[J]. Electric Power Construction,2022,43(2):98-108.
• [9]宋建，束洪春，董俊，等.基于GM(1,1)与BP神经网络的综合负荷预测[J].电力建设，2020,41(5):75-80.SONG Jian,SHU Hongchun,DONG Jun,et al.Comprehensive load forecast based on GM(1,1)and BP neural network[J]. Electric Power Construction,2020,41(5):75-80.
• [10]黄冬梅，庄兴科，胡安铎，等.基于灰色关联分析和K均值聚类的短期负荷预测[J].电力建设，2021,42(7):110-117.HUANG Dongmei,ZHUANG Xingke,HU Anduo,et al.Short-term load forecasting based on similar-day selection with GRA-K-means[J]. Electric Power Construction,2021,42(7):110-117.
• [11]赵允文，李鹏，孙煜皓，等.基于相空间重构和随机配置网络的电力负荷短期预测[J].电力建设，2021,42(9):120-128.ZHAO Yunwen,LI Peng,SUN Yuhao,et al.Short-term power load forecasting based on phase space reconstruction and stochastic configuration networks[J]. Electric Power Construction,2021,42(9):120-128.
• [12]何大四，张旭.改进的季节性指数平滑法预测空调负荷分析[J].同济大学学报（自然科学版），2005,33(12):1672-1676.HE Dasi,ZHANG Xu. Analysis of air conditioning load prediction by modified seasonal exponential smoothing model[J].Journal of Tongji University(Natural Science),2005,33(12):1672-1676.
• [13] ZHAO J,DUAN Y Q,LIU X J.Uncertainty analysis of weather forecast data for cooling load forecasting based on the Monte Carlo method[J].Energies,2018,11(7):1900.
• [14] DYSON M E H,BORGESON S D,TABONE M D,et al. Using smart meter data to estimate demand response potential,with application to solar energy integration[J].Energy Policy,2014,73:607-619.
• [15]方娜，余俊杰，李俊晓，等.注意力机制下的EMD-GRU短期电力负荷预测[J].华侨大学学报（自然科学版），2021,42(6):817-824.FANG Na,YU Junjie,LI Junxiao,et al.Short-term power load forecasting under EMD-GRU attention mechanism[J]. Journal of Huaqiao University(Natural Science),2021,42(6):817-824.
• [16]赵兵，王增平，纪维佳，等.基于注意力机制的CNN-GRU短期电力负荷预测方法[J].电网技术，2019,43(12):4370-4376.ZHAO Bing,WANG Zengping,JI Weijia,et al.A shortterm power load forecasting method based on attention mechanism of CNN-GRU[J].Power System Technology,2019,43(12):4370-4376.
• [17]龚飘怡，罗云峰，方哲梅，等.基于Attention-BiLSTMLSTM神经网络的短期电力负荷预测方法[J].计算机应用，2021,41（增刊1）:81-86.GONG P Y,LUO Y F,FANG Z M,et al. Short-term power load forecasting method based on AttentionBiLSTM-LSTM neural network[J].Journal of Computer Applications,2021,41(S1):81-86.
• [18]赵佩，代业明.基于实时电价和加权灰色关联投影的SVM电力负荷预测[J].电网技术，2020,44(4):1325-1332.ZHAO Pei,DAI Yeming.Power load forecasting of SVM based on real-time price and weighted grey relational projection algorithm[J].Power System Technology,2020,44(4):1325-1332.
• [19]刘倩颖，阮应君，时翔，等.基于k-means聚类与BP神经网络算法的办公建筑逐时电负荷预测[J].热能动力工程，2018,33(3):138-144.LIU Qianying,RUAN Yingjun,SHI Xiang,et al.Hourly power load forecasting of office building based on kmeans clustering and BP neural network algorithm[J].Journal of Engineering for Thermal Energy and Power,2018,33(3):138-144.
• [20]顾春锋，罗其华，奚培锋，等.基于PCA-LSTM神经网络的建筑空调负荷预测方法研究[J].现代建筑电气，2021,12(10):1-7.GU Chunfeng,LUO Qihua,XI Peifeng,et al.Research on load forecasting method of building air conditioning based on PCA-LSTM neural network[J].Modern Architecture Electric,2021,12(10):1-7.
• [21]祁鑫，王福忠，张丽，等.基于SVD-LSTM的高校学生宿舍空调负荷预测[J].电子科技，2020,33(11):59-66.QI Xin,WANG Fuzhong,ZHANG Li,et al.Prediction of air conditioning load in college dormitory based on SVDLSTM[J]. Electronic Science and Technology,2020,33(11):59-66.
• [22]赵超，戴坤成.自适应加权最小二乘支持向量机的空调负荷预测[J].重庆大学学报，2016,39(1):55-64.ZHAO Chao,DAI Kuncheng. Modeling air-conditioning load forecasting based on adaptive weighted least squares support vector machine[J].Journal of Chongqing University,2016,39(1):55-64.
• [23]湛归，殷爽睿，艾芊，等.智能楼宇型虚拟电厂参与电力系统调频辅助服务策略[J].电力工程技术，2022,41(6):13-20.ZHAN Gui,YIN Shuangrui,AI Qian,et al.A strategy for smart building-based virtual power plants participating in frequency regulation auxiliary service[J]. Electric Power Engineering Technology,2022,41(6):13-20.
• [24]雷旭，马鹏飞，宋智帅，等.计及风电预测误差的柔性负荷日内调度模型[J].发电技术，2022,43(3):485-491.LEI Xu,MA Pengfei,SONG Zhishuai,et al.A flexible intraday load dispatch model considering wind power prediction errors[J]. Power Generation Technology,2022,43(3):485-491.
• [25]迟福建，孙阔，张章，等.网格化规划系统中电气计算核心算法的实现[J].发电技术，2022,43(6):860-868.CHI Fujian,SUN Kuo,ZHANG Zhang,et al.Realization of core algorithm of electrical calculation in grid planning system[J].Power Generation Technology,2022,43(6):860-868.
• [26]罗凡，余向前，王林信，等.基于聚类与自适应划分的短期负荷预测[J].电力电容器与无功补偿，2021,42(6):184-189.LUO Fan,YU Xiangqian,WANG Linxin,et al. Shortterm load prediction based on clustering and adaptive partition[J]. Power Capacitor&Reactive Power Compensation,2021,42(6):184-189.
• [27]杨星磊，项川，姜鸣瞻，等.计及产业结构和温度因素的神经网络电网负荷预测方法[J].电力电容器与无功补偿，2022,43(6):79-84.YANG Xinglei,XIANG Chuan,JIANG Mingzhan,et al.Neural network load estimation method considering industrial structure and temperature factors[J].Power Capacitor&Reactive Power Compensation,2022,43(6):79-84.