基于自编码网络和经济增长数据的工业负荷预测模型Industrial Load Forecasting Model Based on Auto-encoder and Economic Growth Rate
孙钢,杨宁,柳文轩,吴磊,韩蕾,赵俊华
SUN Gang,YANG Ning,LIU Wenxuan,WU Lei,HAN Lei,ZHAO Junhua
摘要(Abstract)：
考虑到产业发展与用电量之间的显著相关性,采用自动编码器网络这一深度学习方法来研究电力消费与经济发展之间的关系。研究了如何根据经济增长数据预测工业用电量。算例试验研究结果表明:自动编码器网络算法在大多数情况下比一般的线性ARMA方法具有更高的精度;统计上,第二产业几乎所有行业都高度依赖电力消费,其电力消费与其增长率呈非线性正相关。
Considering the significant correlation between industrial development and electricity consumption,this paper employs auto-encoder a deep learning method, to study the relationship between electricity consumption and economic development. This paper also investigates how to forecast the industrial electricity consumption based on economic growth data. The case study show that the AE algorithm demonstrates higher accuracies in most cases than common linear ARMA method; statistically, almost all the sectors in the secondary industry are highly dependent on the electricity consumption, and their electricity consumption is nonlinearly and positively correlated with their growth rate.
关键词(KeyWords)：
行业电耗;第二产业;经济增长;自编码网络;预测
electric consumption;secondary industry;economic growth;auto-encoder;forecasting
基金项目(Foundation): 深圳市科技创新委员会支持项目(GJHZ20160301165723718,JCYJ20170410172224515);; 浙江华云信息科技有限公司科技项目
作者(Author):
孙钢,杨宁,柳文轩,吴磊,韩蕾,赵俊华
SUN Gang,YANG Ning,LIU Wenxuan,WU Lei,HAN Lei,ZHAO Junhua
DOI: 10.19585/j.zjdl.201808003
参考文献(References)：
- [1]BURNS A,MITCHELL W.Measuring business cycles[M].Nber Books,1946:70-71.
- [2]HAMILTON J.A new approach to the economic analysis of nonstationary time series and the business cycle[J].E-conometrica:Journal of the Econometric Society,1989,4(1):357-384.
- [3]陈磊,高铁梅.利用Stock-Watson型景气指数对宏观经济形势的分析和预测[J].数量经济技术经济研究,1994,5(5):53-59.
- [4]STUDENMUND A H,HENRY J C.Using Econometrics:Apractical guide[M].New Jersey:Addison-Wesley Educational Publishers,1992:2,163,215.
- [5]YANN L,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521.7553:436.
- [6]BENGIO Y.Learning deep architectures for AI[J].Foundations and Trends in Machine Learning,2009,2(1):1-127.
- [7]梁亚丽.中国电力需求周期波动特征及影响因素研究[D].北京:华北电力大学,2005.
- [8]吴疆.电力消费与宏观经济的相关性研究[J].中国能源,2009,31(6):17-21.
- [9]中国电力企业联合会.2009年上半年全国电力供需与经济运行形势分析[J].中国电力发展与改革研究,2009(6):35-40.
- [10]陈文静,何刚.电力消费及其影响因素:基于非参数模型的研究[J].系统工程理论与实践,2009,29(5):92-97.
- [11]HINTON G,RICHARD Z.Autoencoders,minimum descrip tion length and Helmholtz free energy[C]//Advances in neural information processing systems,1993,6(1):3-10.
- [12]MAKHZANI,ALIREZA,BRENDAN F.K-sparse autoencoders[J].ar Xiv preprint ar Xiv:1312.5663,2013.
- [13]VINCENT P,LAROCHELLE H,BENGIO Y,et al.Extracting and composing robust features with denoising autoen coders[C]//International Conference on Machine Learning.ACM,2008:1096-1103.
- [14]HONG C,CHEN X,WANG X,et al.Hypergraph regularized autoencoder for image-based 3D human pose recovery[J].Signal Processing,2016,124(1):132-140.
- [15]MESNIL G,DAUPHIN Y,GLOROT X,et al.Unsupervised and Transfer Learning Challenge:a Deep Learning Approach[J].Workshop on Unsupervised&Transfer Learning,2012(7):1-15.
- [16]CHEN M,XU Z,WEINBERGER K,et al.Marginalized Denoising Autoencoders for Domain Adaptation[J].Computer Science,2012.
- [17]BETECHUOH B L,MARWALA T,TETTEY T.Autoencoder networks for HIV classification[J].Current Science,2006,91(11):1467-1473.
- [18]FENG F,WANG X,LI R.Cross-modal retrieval with correspondence autoencoder[C]//Proceedings of the 22nd ACMinternational conference on Multimedia.ACM,2014:7-16.
- [19]LANGE S,RIEDMILLER M.Deep auto-encoder neural networks in reinforcement learning[C]//International Joint Conference on Neural Networks.IEEE,2010:1-8.
- [20]TAO D,HONG C,YU J,et al.Multimodal Deep Autoencoder for Human Pose Recovery[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,2015,24(12):5659-5670.
- [21]LI J,LUONG M T,JURAFSKY D.A Hierarchical Neural Autoencoder for Paragraphs and Documents[J].Computer Science,2015.
- [22]GREENE W H.On the Asymptotic Bias of the Ordinary Least Squares Estimator of the Tobit Model[J].Econometrica,1981,49(2):505-513.
- [23]HOLLANDER M,WOLFE D A.Nonparametric statistical methods[M].Robust nonparametric statistical methods.Arnold,1998.