基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测

doi:10.11772/j.issn.1001-9081.2020060929

计算机应用 ›› 2021, Vol. 41 ›› Issue (1): 231-236.DOI: 10.11772/j.issn.1001-9081.2020060929

所属专题：第八届中国数据挖掘会议(CCDM 2020)

• 第八届中国数据挖掘会议(CCDM 2020) • 上一篇下一篇

基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测

肖勇¹, 郑楷洪¹, 郑镇境², 钱斌¹, 李森², 马千里²

1. 南方电网科学研究院有限责任公司, 广州 510080;
2. 华南理工大学计算机科学与工程学院, 广州 510006

收稿日期:2020-05-31 修回日期:2020-07-20 出版日期:2021-01-10 发布日期:2020-11-12
通讯作者: 马千里
作者简介:肖勇(1978-),男,湖南怀化人,高级工程师,博士,主要研究方向:电能计量管理、电能计量自动化系统、用电技术;郑楷洪(1991-),男,广东汕头人,工程师,硕士,主要研究方向:电能计量、电能计量自动化系统、用电技术;郑镇境(1996-),男,广东揭阳人,硕士研究生,主要研究方向:数据挖掘、机器学习、神经网络;钱斌(1989-),男,湖北十堰人,工程师,硕士,主要研究方向:电能计量;李森(1994-),男,广东茂名人,硕士研究生,主要研究方向:数据挖掘、机器学习、神经网络;马千里(1980-),男,甘肃宕昌人,教授,博士,主要研究方向:数据挖掘、机器学习、神经网络。
基金资助:
国家自然科学基金重点项目（61751205）；国家自然科学基金资助项目（61872148）。

Multi-scale skip deep long short-term memory network for short-term multivariate load forecasting

XIAO Yong¹, ZHENG Kaihong¹, ZHENG Zhenjing², QIAN Bin¹, LI Sen², MA Qianli²

1. Electric Power Research Institute, China Southern Power Grid International Company Limited, Guangzhou Guangdong 510080, China;
2. School of Computer Science and Engineering, South China University of Technology, Guangzhou Guangdong 510006, China

Received:2020-05-31 Revised:2020-07-20 Online:2021-01-10 Published:2020-11-12
Supported by:
This work is partially supported by the Key Project of National Natural Science Foundation of China (61751205), the National Natural Science Foundation of China (61872148).

摘要/Abstract

摘要： 近年来，以循环神经网络（RNN）为主体构建的预测模型在短期电力负荷预测中取得了优越的性能。然而，由于RNN不能有效捕捉存在于短期电力负荷数据的多尺度时序特征，因而难以进一步提升负荷预测精度。为了捕获短期电力负荷数据中的多尺度时序特征，提出了一种基于多尺度跳跃深度长短期记忆（MSD-LSTM）网络的短期电力负荷预测模型。具体来说，以长短期记忆（LSTM）网络为主体构建预测模型能够较好地捕获长短期时序依赖，从而缓解时序过长时重要信息容易丢失的问题。进一步地，采用多层LSTM架构并且对各层设置不同的跳跃连接数，使得MSD-LSTM的每一层能够捕获不同时间尺度的特征。最后，引入全连接层把各层提取到的多尺度时序特征进行融合，再利用该融合特征进行短期电力负荷预测。实验结果表明，与单层LSTM和多层LSTM相比，MSD-LSTM的均方误差总体下降了10%。可见MSD-LSTM能够更好地提取短期负荷数据中的多尺度时序特征，从而提高短期电力负荷预测的精度。

关键词: 短期电力负荷预测, 时间序列预测, 多尺度时序特征, 长短期记忆网络, 跳跃连接

Abstract: In recent years, the short-term power load prediction model built with Recurrent Neural Network (RNN) as main part has achieved excellent performance in short-term power load forecasting. However, RNN cannot effectively capture the multi-scale temporal features in short-term power load data, making it difficult to further improve the load forecasting accuracy. To capture the multi-scale temporal features in short-term power load data, a short-term power load prediction model based on Multi-scale Skip Deep Long Short-Term Memory (MSD-LSTM) was proposed. Specifically, a forecasting model was built with LSTM (Long Short-Term Memory) as main part, which was able to better capture long short-term temporal dependencies, thereby alleviating the problem that important information is easily lost when encountering the long time series. Furthermore, a multi-layer LSTM architecture was adopted and different skip connection numbers were set for the layers, enabling different layers of MSD-LSTM can capture the features with different time scales. Finally, a fully connected layer was introduced to fuse the multi-scale temporal features extracted by different layers, and the obtained fusion feature was used to perform the short-term power load prediction. Experimental results show that compared with LSTM, MSD-LSTM achieves lower Mean Square Error (MSE) with the reduction of 10% in general. It can be seen that MSD-LSTM can better capture multi-scale temporal features in short-term power load data, thereby improving the accuracy of short-term power load forecasting.

Key words: short-term power load forecasting, time series forecasting, multi-scale temporal feature, Long Short-Term Memory (LSTM) network, skip connection

中图分类号:

TM714

肖勇, 郑楷洪, 郑镇境, 钱斌, 李森, 马千里. 基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测[J]. 计算机应用, 2021, 41(1): 231-236.

XIAO Yong, ZHENG Kaihong, ZHENG Zhenjing, QIAN Bin, LI Sen, MA Qianli. Multi-scale skip deep long short-term memory network for short-term multivariate load forecasting[J]. Journal of Computer Applications, 2021, 41(1): 231-236.

参考文献

[1] 崔和瑞, 彭旭. 基于ARIMAX模型的夏季短期电力负荷预测[J]. 电力系统保护与控制,2015,43(4):108-114.(CUI H R, PENG X. Summer short-term load forecasting based on ARIMAX model[J]. Power System Protection and Control,2015,43(4):108-114.)
[2] DUDEK G. Pattern-based local linear regression models for shortterm load forecasting[J]. Electric Power Systems Research,2016, 130:139-147.
[3] TAKEDA H,TAMURA Y,SATO S. Using the ensemble Kalman filter for electricity load forecasting and analysis[J]. Energy, 2016,104:184-198.
[4] 吴云, 雷建文, 鲍丽山, 等. 基于改进灰色关联分析与蝙蝠优化神经网络的短期负荷预测[J]. 电力系统自动化,2018,42(20):67-72.(WU Y,LEI J W,BAO L S,et al. Short-term load forecasting based on improved grey relational analysis and neural network optimized by bat algorithm[J]. Automation of Electric Power Systems,2018,42(20):67-72.)
[5] 罗育辉, 蔡延光, 戚远航, 等. 基于最大偏差相似性准则的BP神经网络短期电力负荷预测算法[J]. 计算机应用研究,2019,36(11):3269-3273.(LUO Y H,CAI Y G,QI Y H,et al. Short-term power load forecasting algorithm based on maximum deviation similarity criterion BP neural network[J]. Application Research of Computers,2019,36(11):3269-3273.)
[6] FAN G,PENG L,HONG W C,et al. Electric load forecasting by the SVR model with differential empirical mode decomposition and auto regression[J]. Neurocomputing,2016,173(Pt 3):958-970.
[7] 任利强, 张立民, 王海鹏, 等. 基于IPSO-GPR的短期负荷区间预测[J]. 计算机工程与设计,2019,40(10):3002-3008.(REN L Q,ZHANG L M,WANG H P,et al. Short-term power load interval forecasting based on improved particle swarm optimization and Gaussian process regression[J]. Computer Engineering and Design,2019,40(10):3002-3008.)
[8] ELMAN J L. Finding structure in time[J]. Cognitive Science, 1990,14(2):179-211.
[9] HOCHREITER S,SCHMIDHUBER J. Long short-term memory[J]. Neural Computation,1997,9(8):1735-1780.
[10] ZHENG J,XU C,ZHANG Z,et al. Electric load forecasting in smart grids using long-short-term-memory based recurrent neural network[C]//Proceedings of the 51st Annual Conference on Information Sciences and Systems. Piscataway:IEEE,2017:1-6.
[11] 李鹏, 何帅, 韩鹏飞, 等. 基于长短期记忆的实时电价条件下智能电网短期负荷预测[J]. 电网技术,2018,42(12):4045-4052.(LI P,HE S,HAN P F,et al. Short-term load forecasting of smart grid based on long-short-term memory recurrent neural networks in condition of real-time electricity price[J]. Power System Technology,2018,42(12):4045-4052.)
[12] 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 C Q,XIA Q,ZHANG B M. Review of power system load forecasting and its development[J]. Automation of Electric Power Systems, 2004,28(17):1-11.)
[13] MA Q,CHEN E,LIN Z,et al. Convolutional multitimescale echo state network[J]. IEEE Transactions on Cybernetics,2019(Early Access):1-13.
[14] MA Q,ZHENG J,LI S,et al. Learning representations for time series clustering[C]//Proceedings of the 33rd Annual Conference on Neural Information Processing Systems. Red Hook, NY:Curran Associates Inc.,2019:3776-3786.
[15] CHANG S,ZHANG Y,HAN W,et al. Dilated recurrent neural networks[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook,NY:Curran Associates Inc.,2017:77-87.
[16] LILLICRAP T P,SANTORO A. Backpropagation through time and the brain[J]. Current Opinion in Neurobiology,2019,55:82-89.
[17] SMOLA A J, SCHÖLKOPF B. A tutorial on support vector regression[J]. Statistics and Computing,2004,14(3):199-222.
[18] RASMUSSEN C E. Gaussian processes in machine learning[C]//Proceedings of the 2003 Summer School on Machine Learning, LNCS 3176. Berlin:Springer,2003:63-71.

基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测

Multi-scale skip deep long short-term memory network for short-term multivariate load forecasting

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	李康康, 张静. 基于注意力机制的多层次编码和解码的图像描述模型[J]. 计算机应用, 2021, 41(9): 2504-2509.
[2]	党伟超, 李涛, 白尚旺, 高改梅, 刘春霞. 基于自注意力长短期记忆网络的Web软件系统实时剩余寿命预测方法[J]. 计算机应用, 2021, 41(8): 2346-2351.
[3]	武国亮, 徐继宁. 基于命名实体识别任务反馈增强的中文突发事件抽取方法[J]. 计算机应用, 2021, 41(7): 1891-1896.
[4]	张元钧, 张曦煌. 基于图卷积与长短期记忆网络的动态网络表示学习模型[J]. 计算机应用, 2021, 41(7): 1857-1864.
[5]	李洋莹, 陈智军, 张子豪, 游兰. 基于改进Elman神经网络的制糖企业原糖需求预测模型[J]. 计算机应用, 2021, 41(7): 2113-2120.
[6]	武光利, 李雷霆, 郭振洲, 王成祥. 基于改进的双向长短期记忆网络的视频摘要生成模型[J]. 计算机应用, 2021, 41(7): 1908-1914.
[7]	徐萌, 王亚锟. 基于双向长短期记忆网络的DA40飞机碳刹车片剩余寿命预测[J]. 计算机应用, 2021, 41(5): 1527-1532.
[8]	李文惠, 曾上游, 王金金. 基于改进注意力机制的图像描述生成算法[J]. 计算机应用, 2021, 41(5): 1262-1267.
[9]	张欣环, 刘宏杰, 施俊庆, 毛程远, 孟国连. 基于时空特征向量的长短期记忆人工神经网络的城市公交旅行时间预测[J]. 计算机应用, 2021, 41(3): 875-880.
[10]	许力, 李建华. 基于句法依存分析的图网络生物医学命名实体识别[J]. 计算机应用, 2021, 41(2): 357-362.
[11]	钱斌, 郑楷洪, 陈子鹏, 肖勇, 李森, 叶纯壮, 马千里. 基于残差连接长短期记忆网络的时间序列修复模型[J]. 计算机应用, 2021, 41(1): 243-248.
[12]	傅魁, 梁少晴, 李冰. 基于改进的深度Q网络结构的商品推荐模型[J]. 计算机应用, 2020, 40(9): 2613-2621.
[13]	林志兴, 王立可. 基于深度特征和Seq2Seq模型的网络态势预测方法[J]. 计算机应用, 2020, 40(8): 2241-2247.
[14]	胡学敏, 童秀迟, 郭琳, 张若晗, 孔力. 基于深度视觉注意神经网络的端到端自动驾驶模型[J]. 计算机应用, 2020, 40(7): 1926-1931.
[15]	代强, 程曦, 王永梅, 牛子未, 刘飞. 基于轻量自动残差缩放网络的图像超分辨率重建[J]. 计算机应用, 2020, 40(5): 1446-1452.