基于MM-LSTM的非工楼宇中央空调负荷预测
Central Air Conditioning Load Forecasting for Non-Industrial Buildings Based on MM-LSTM
DOI: 10.12677/SG.2021.113027, PDF,   
作者: 杨 斌, 陈振宇, 霍 尧, 阮文骏：国网江苏电力有限公司，江苏 南京；罗玉孙：北京慧和仕科技有限责任公司，北京
关键词: 中央空调负荷预测；多路多层LSTM；神经网络；非工楼宇；Central Air Conditioning Load Forecasting； MM-LSTM； Neural Network； Non-Industrial Buildings
摘要: 空调负荷已经成为电网负荷中非常重要的组成部分，特别是在非工楼宇空调，对非工楼宇空调负荷进行准确负荷预测，可以进行更加合理的需求响应，提高区域用电质量。MM-LSTM (多路多层LSTM网络)预测方法对非工楼宇中央空调的用电负荷进行预测精度较高。通过数据实验表明，该负荷预测方法在适当的数据特征工程后，预测普遍优于一般LSTM网络的空调负荷预测方法，具有良好的预测精度。该方法可有效满足非工楼宇空调需求响应的需要，具有良好的实用价值。
Abstract: Air-conditioning load has become a very important part of the grid load, especially in non-industrial buildings. Accurate load forecasting of non-industrial building air-conditioning loads can perform more reasonable demand response and improve regional power quality. MM-LSTM (multi-channel & multi-layer LSTM network) prediction method is used to predict the load of non-industrial building air conditioners in a short time, and the high-precision building air conditioning load prediction results are obtained. The experimental data shows that the proposed method is superior to the general LSTM network method under the premise of suitable feature engineering, and has good prediction accuracy. The method can effectively meet the needs of non-industrial building air conditioning demand response and has good practical value.
文章引用：杨斌, 陈振宇, 霍尧, 阮文骏, 罗玉孙. 基于MM-LSTM的非工楼宇中央空调负荷预测[J]. 智能电网, 2021, 11(3): 288-295. https://doi.org/10.12677/SG.2021.113027

参考文献

[1] 杨永标, 颜庆国, 徐石明, 肖文举, 薛铭枫. 公共楼宇空调负荷参与电网虚拟调峰的思考[J]. 电力系统自动化, 2015, 39(17): 103-107.
[2] 王怡岚, 童亦斌, 黄梅, 杨利, 赵洪刚. 基于需求侧响应的空调负荷虚拟储能模型研究[J]. 电网技术, 2017, 41(2): 394-401.
[3] Forrester, R.J. and Wepfer, W.J. (1984) Formulation of Prediction Algorithms for Management of Commercial and Industrial Energy Loads. ASHRAE Transactions, 85, 234-239.
[4] 李爱旗. 重庆市小城镇居住建筑热环境分析和建筑冷热负荷预测研究[D]: [硕士学位论文]. 重庆: 重庆大学, 2006.
[5] 谢艳群. 长沙市住宅建筑能耗调查实测及其影响因素统计分析[D]: [硕士学位论文]. 长沙: 湖南大学, 2007.
[6] Macarthur, J.W., Mathur, A. and Zhao, J. (1989) On-Line Recursive Estimation for Load Profile Prediction. ASHRAE Transactions, 95, 621-628.
[7] 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, 2004, 33(12): 1672-1676.
[8] 孙靖, 程大章. 基于季节性时间序列模型的空调负荷预测[J]. 电工技术学报, 2004, 19(3): 88-93.
[9] 孙光伟, 索胜军, 陶永纯, 等. 基于BP网络的空调负荷预报建模方法的研究[J]. 哈尔滨建筑大学学报, 1999, 32(6): 79-82.
[10] 石磊, 赵蕾, 王军, 等. 应用人工神经网络预测建筑物空调负荷[J]. 暖通空调, 2003, 33(1): 103-104, 113.
[11] 王蕾, 张九根, 李腾, 等. 基于改进BP神经网络的中央空调冷负荷预测研究[J]. 计算机测量与控制, 2014, 22(6): 1690-1692.
[12] Vermaak, J. and Botha, E.C. (1998) Recurrent Neural Networks for Short-Term Load Forecasting. IEEE Transactions on Power Systems, 13, 126-132.
[Google Scholar] [CrossRef] 
[13] Marino, D.L., Amarasinghe, K. and Manic, M. (2016) Building Energy Load Forecasting Using Deep Neural. IECON 2016: 42nd Annual Conference of the IEEE Industrial Electronics Society, Florence, 23-26 October 2016 7046-7051.
[Google Scholar] [CrossRef] 
[14] Hochreiter, S. and Schmidhuber, J. (1997) Long Short-Term Memory. Neural Computation, 9, 1735-1780.
[Google Scholar] [CrossRef] [PubMed]
[15] Gers, F.A., Schmidhuber, J. and Cummins, F. (2000) Learning to Forget: Continual Prediction with LSTM. Neural computation, 12, 2451-2471.
[Google Scholar] [CrossRef] [PubMed]
[16] Gers, F.A., Schraudolph, N. and Schmidhuber, J. (2003) Learning Precise Timing with LSTM Recurrent Networks. The Journal of Machine Learning Research, 3, 115-143.
[17] Graves, A. (2012) Supervised Sequence Labelling with Recurrent Neural Networks. Vol. 385, Springer, Berlin.
[Google Scholar] [CrossRef] 
[18] 刘学之, 杨泽宇, 沈凤武, 尚胡佟, 刘嘉. 基于S型曲线的指标非线性标准化研究[J]. 统计与信息论坛, 2018, 33(2): 17-21.

为你推荐