姚程文;杨苹;刘泽健;

• 1:华南理工大学电力学院
• 2:广东省绿色能源技术重点实验室(华南理工大学)

摘要(Abstract)：

为充分挖掘负荷数据中时序性特征的联系,提高负荷预测的精度,提出了一种基于卷积神经网络(convolutional neural networks,CNN)和门控循环单元(gated recurrent unit,GRU)混合神经网络的负荷预测方法。以日期因素、气候因素、相似日负荷因素构建特征集作为输入,首先采用k-means聚类方法对地区内的样本数据集进行分组;再运用CNN网络提取特征与负荷在高维空间的联系,构造时序序列的高维特征向量,并将结果输入到GRU网络中;最后训练各组GRU网络模型的参数并输出负荷预测值。使用该方法对浙江省某地区电力负荷数据进行预测,结果表明,所提负荷预测方法与长短期记忆(long short-termmemory,LSTM)网络模型、GRU网络模型、CNN-LSTM网络模型、支持向量机回归模型及决策树模型相比,在预测精度与预测效率方面具有显著优势。

关键词(KeyWords)： 负荷预测;卷积神经网络;门控循环单元;深度学习;负荷聚类

Abstract:

Keywords:

基金项目(Foundation): 广东省科技计划项目(2017B030314124)~~

作者(Author): 姚程文;杨苹;刘泽健;

Email:

DOI: 10.13335/j.1000-3673.pst.2019.2058

参考文献(References)：

• [1]肖白，黎平．城网空间电力负荷预测中的负荷规律性分析[J]．电网技术，2009,33(20):113-119.Xiao Bai,Li Ping.Analysis of load regularity in urban power network space power load forecasting[J].Power System Technology,2009,33(20):113-119(in Chinese).
• [2]康重庆，夏清，张伯明．电力系统负荷预测研究综述与发展方向的探讨[J]．电力系统自动化，2004,28(17):1-11.Kang Chongqing,Xia Qing,Zhang Boming.Review of power system load forecasting and discussion on development trends[J].Automation of Electric Power Systems,2004,28(17):1-11(in Chinese).
• [3]肖白，周潮，穆钢．空间电力负荷预测方法综述与展望[J]．中国电机工程学报，2013,33(25):78-92,14.Xiao Bai,Zhou Chao,Mu Gang.Overview and prospect of space power load forecasting methods[J].Proceedings of the CSEE,2013,33(25):78-92,14(in Chinese).
• [4]赵峰，孙波，张承慧．基于多变量相空间重构和卡尔曼滤波的冷热电联供系统负荷预测方法[J]．中国电机工程学报，2016,36(2):399-406.Zhao Feng,Sun Bo,Zhang Chenghui.Load forecasting method of combined cooling,heating and power system based on multivariable phase space reconstruction and Kalman filtering[J].Proceedings of the CSEE,2016,36(2):399-406(in Chinese).
• [5]毛李帆，江岳春，龙瑞华，等．基于偏最小二乘回归分析的中长期电力负荷预测[J]．电网技术，2008,36(19):71-77.Mao Lifan,Jiang Yuechun,Long Ruihua,et al.Mid-and long-term power load forecasting based on partial least squares regression analysis[J].Power System Technology,2008,36(19):71-77(in Chinese).
• [6]谢毓广，刘童，陈凡，等．基于置信度理论和多元线性回归方程的电网负荷预测[J]．电工技术，2018(13):42-45.Xie Yuguang,Liu Tong,Chen Fan,et al.Power network load forecasting based on confidence theory and multiple linear regression equations[J].Electrical Engineering Technology,2018(13):42-45(in Chinese).
• [7]胡海琴，蒋传文，蔡宏欣．中期负荷预测的傅里叶级数残差修正模型[J]．安徽电力，2010,27(2):39-41.Hu Haiqin,Jiang Chuanwen,Cai Hongxin.Fourier series residual correction model for medium-term load forecasting[J].Anhui Electric Power,2010,27(2):39-41(in Chinese).
• [8]吴蓓，丁明，陈闽江．基于模糊推理和多目标规划的空间负荷预测[J]．电网技术，2004,28(15):48-52.Wu Bei,Ding Ming,Chen Minjiang.Spatial load forecasting based on fuzzy inference and multi-objective planning[J].Power System Technology,2004,28(15):48-52(in Chinese).
• [9]陈宇杰．基于BP神经网络的区域配电网中期电力负荷预测[D]．长春：吉林大学，2017.
• [10]王军．基于深度神经网络的中期电力负荷预测[J]．重庆工商大学学报(自然科学版)，2018,35(6):17-21.Wang Jun.Mid-term power load forecasting based on deep neural network[J]. Journal of Chongqing Technology and Business University(Natural Science Edition),2018,35(6):17-21(in Chinese).
• [11]谷云东，马冬芬，程红超．基于相似数据选取和改进梯度提升决策树的电力负荷预测[J]．电力系统及其自动化学报，2019,31(5):64-69.Gu Yundong,Ma Dongfen,Cheng Hongchao.Power load forecasting based on similar data selection and improved gradient boosting decision tree[J].Journal of Electric Power System and Automation,2019,31(5):64-69(in Chinese).
• [12]谢敏，邓佳梁，吉祥，等．基于信息熵和变精度粗糙集优化的支持向量机降温负荷预测方法[J]．电网技术，2017,41(1):210-214.Xie Min,Deng Jialiang,Jixiang,et al.Support vector machine cooling load forecasting method based on information entropy and variable precision rough set optimization[J].Power System Technology,2017,41(1):210-214(in Chinese).
• [13]翟永杰，王静娴，周黎辉．基于模糊支持向量机的电力系统中期负荷预测[J]．华北电力大学学报(自然科学版)，2008,35(2):70-73.Zhai Yongjie,Wang Jingxian,Zhou Lihui.Mid-term load forecasting of power system based on fuzzy support vector machine[J].Journal of North China Electric Power University(Natural Science Edition),2008,35(2):70-73(in Chinese).
• [14]张籍，薛儒涛，刘慧，等．基于深度信念网络的不同行业中长期负荷预测[J/OL]．电力系统及其自动化学报，DOI:10.19635/j.cnki.csu-epsa.000117.Zhang Ji,Xue Rutao,Liu Hui,et al.Medium and long-term load forecasting in different industries based on deep belief networks[J/OL].Journal of Electric Power Systems and Automation,DOI:10.19635/j.cnki.csu-epsa.000117(in Chinese).
• [15]周中明，李建平，张鑫，等．ANN在电力系统中期负荷预测中的应用[J]．电力系统及其自动化学报，2003,15(2):36-39.Zhou Zhongming,Li Jianping,Zhang Xin,et al.Application of ANN in medium-term load forecasting of power system[J].Journal of Electric Power System and Automation,2003,15(2):36-39(in Chinese).
• [16]彭文，王金睿，尹山青．电力市场中基于Attention-LSTM的短期负荷预测模型[J]．电网技术，2019,43(5):1745-1751.Peng Wen,Wang Jinrui,Yin Shanqing.Attention-LSTM-based short-term load forecasting model in power market[J].Power System Technology,2019,43(5):1745-1751(in Chinese).
• [17]周莽，高僮，李晨光．等．GRU神经网络短期电力负荷预测研究[J]．科技创新与应用，2018(33):52-53,57.Zhou Mang,Gao Tong,Li Chenguang,et al.Research on power load forecasting based on GRU neural network[J].Science&Technology Innovation and Application,2018(33):52-53,57(in Chinese).
• [18]陈海文，王守相，王绍敏，等．基于门控循环单元网络与模型融合的负荷聚合体预测方法[J]．电力系统自动化，2019,43(1):65-74.Chen Haiwen,Wang Shouxiang,Wang Shaomin,et al.Load aggregation forecasting method based on gated recurrent unit network and model fusion[J].Automation of Electric Power Systems,2019,43(1):65-74(in Chinese).
• [19]陆继翔，张琪培，杨志宏，等．基于CNN-LSTM混合神经网络模型的短期负荷预测方法[J]．电力系统自动化，2019,43(8):131-137.Lu Jixiang,Zhang Qipei,Yang Zhihong,et al.Short-term load forecasting method based on CNN-LSTM hybrid neural network model[J].Automation of Electric Power Systems,2019,43(8):131-137(in Chinese).
• [20]姜新猛．基于TensorFlow的卷积神经网络的应用研究[D]．武汉：华中师范大学，2017.
• [21] Cho K,Merrienboer B V,Bahdanau D,et al.On the properties of neural machine translation:encoder-decoder approaches[EB/OL].[2018-06-05].https://arxiv.org/pdf/1409.1259.pdf.
• [22]陈希，李玲娟．基于降维和聚类的协同过滤推荐算法[J]．计算机技术与发展，2020(1):1-5.Chen Xi, Li Lingjuan. Collaborative filtering recommendation algorithm based on dimensionality reduction and clustering[J].Computer Technology and Development,2020(1):1-5(in Chinese).
• [23]陈土生．基于Adam优化GRU神经网络的SCADA系统入侵检测方法[J]．现代计算机，2019(15):13-19.Chen Tusheng.Intrusion detection method of SCADA system based on Adam’s optimized GRU neural network[J].Modern Computer,2019(15):13-19(in Chinese).
• [24]刘俊，赵宏炎，刘嘉诚，等．基于协整-格兰杰因果检验和季节分解的中期负荷预测[J]．电力系统自动化，2019,43(1):73-80.Liu Jun,Zhao Hongyan,Liu Jiacheng,et al.Mid-term load forecasting based on co-integration-granger causality test and seasonal decomposition[J].Automation of Electric Power Systems,2019,43(1):73-80(in Chinese).
• [25]吴广建，章剑林，袁丁．基于K-means的手肘法自动获取K值方法研究[J]．软件，2019,40(5):167-170.Wu Guangjian, Zhang Jianlin, Yuan Ding. Research on K-means-based elbow method to automatically obtain K value[J].Software,2019,40(5):167-170(in Chinese).