A hybrid short-term load forecasting method based on improved ensemble empirical mode decomposition and back propagation neural network

Short-term load forecasting (STLF) plays a very important role in improving the economy and security of electricity system operations. In this paper, a hybrid STLF method is proposed based on the improved ensemble empirical mode decomposition (IEEMD) and back propagation neural network (BPNN). To alleviate the mode mixing and end-effect problems in traditional empirical mode decomposition (EMD), an IEEMD is presented based on the degree of wave similarity. By applying the IEEMD method, the nonlinear and nonstationary original load series is decomposed into a finite number of stationary intrinsic mode functions (IMFs) and a residual. Among these components, the high frequency (namely IMF1) is always so small that it has little contribution to model fitting, while it sometimes has a great disturbance for the STLF. Therefore, the IMF1 is removed in the proposed hybrid method for denoising. The remaining IMFs and residual are forecast by BPNN, and then the forecasting results of each component are combined with BPNN to obtain the final predicted load series. Three groups of studies were done to evaluate the effectiveness of the proposed hybrid method. The results show that the proposed hybrid method outperforms other methods both mentioned in this paper and previous studies in terms of all the three standard statistical indicators considered in this study.中文概要目 的短期电力负荷预测是电力系统安全调度、经济运 行的重要依据。研究处理非线性、非稳态电力负 荷信号的新方法, 建立短期负荷预测的混合模 型, 提高短期负荷预测的精确度。创新点1. 提出一种改进总体经验模态分解(EEMD)方 法, 抑制传统EEMD 方法中的端点效应问题; 2. 提出一种基于改进EEMD 和反向传播神经网 络(BPNN)的短期负荷预测方法。方 法1. 使用改进的EEMD 方法将非稳态、非线性的电 力负荷信号分解为一系列的内禀模态函数和一 个趋势余量; 2. 移除所得的高频内禀模态函数; 3. 使用BPNN分别预测各内禀模态函数及趋势余 量; 4. 使用BPNN 组合各内禀模态函数及趋势余 量预测结果, 即为最终负荷预测结果。结 论1. 所提出的改进EEMD 方法能有效抑制传统 EEMD方法中的端点效应问题; 2. 在相同条件下, 所提出的基于改进EEMD 和BPNN 的短期负荷 预测方法较 BPNN、EMD-BPNN、EEMD-BPNN、 SARIMA-BPNN、WTNNEA 和WGMIPSO 预测 方法有更高的精确度。

[1]  G. Sudheer,et al.  Short term load forecasting using wavelet transform combined with Holt–Winters and weighted nearest neighbor models , 2015 .

[2]  K. Chau,et al.  Improving forecasting accuracy of medium and long-term runoff using artificial neural network based on EEMD decomposition. , 2015, Environmental research.

[3]  Jaime Lloret,et al.  Short-Term Load Forecasting for Microgrids Based on Artificial Neural Networks , 2013 .

[4]  Rajesh Patel,et al.  Denoising of multichannel MCG data by the combination of EEMD and ICA and its effect on the pseudo current density maps , 2015, Biomed. Signal Process. Control..

[5]  Rahmat-Allah Hooshmand,et al.  Short term electric load forecasting by wavelet transform and grey model improved by PSO (particle swarm optimization) algorithm , 2014 .

[6]  A. Goia,et al.  Functional clustering and linear regression for peak load forecasting , 2010 .

[7]  Jie Wu,et al.  Short term load forecasting technique based on the seasonal exponential adjustment method and the regression model , 2013 .

[8]  Jianzhou Wang,et al.  An adaptive fuzzy combination model based on self-organizing map and support vector regression for electric load forecasting , 2012 .

[9]  Zhiheng Li,et al.  Multiple-step ahead Traffic Forecasting based on GMM Embedded BP Network , 2013 .

[10]  C. Y. Yang,et al.  Diagnostics of gear deterioration using EEMD approach and PCA process , 2015 .

[11]  Shengzhi Huang,et al.  Monthly streamflow prediction using modified EMD-based support vector machine , 2014 .

[12]  Hui Liu,et al.  An EMD-recursive ARIMA method to predict wind speed for railway strong wind warning system , 2015 .

[13]  Zhongyi Hu,et al.  Interval Forecasting of Electricity Demand: A Novel Bivariate EMD-based Support Vector Regression Modeling Framework , 2014, ArXiv.

[14]  Norden E. Huang,et al.  Ensemble Empirical Mode Decomposition: a Noise-Assisted Data Analysis Method , 2009, Adv. Data Sci. Adapt. Anal..

[15]  Chris Chatfield,et al.  What is the ‘best’ method of forecasting? , 1988 .

[16]  T. Funabashi,et al.  Next day load curve forecasting using hybrid correction method , 2005, IEEE Transactions on Power Systems.

[17]  Hu Chen,et al.  Nonlinear response prediction of cracked rotor based on EMD , 2013, J. Frankl. Inst..

[18]  Caihong Li,et al.  A New Strategy for Short-Term Load Forecasting , 2013 .

[19]  Tao Chen,et al.  Back propagation neural network with adaptive differential evolution algorithm for time series forecasting , 2015, Expert Syst. Appl..

[20]  Saifur Rahman,et al.  An expert system based algorithm for short term load forecast , 1988 .

[21]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[22]  Yi Yang,et al.  A hybrid application algorithm based on the support vector machine and artificial intelligence: An example of electric load forecasting , 2015 .

[23]  D.W. Bunn,et al.  Forecasting loads and prices in competitive power markets , 2000, Proceedings of the IEEE.

[24]  Haiyan Lu,et al.  Multi-step forecasting for wind speed using a modified EMD-based artificial neural network model , 2012 .

[25]  Haibin Wang,et al.  Using AMR data for load estimation for distribution system analysis , 2006 .

[26]  Gwilym M. Jenkins,et al.  Some practical aspects of forecasting in organizations , 1982 .

[27]  Dan J. Trudnowski,et al.  Real-time very short-term load prediction for power-system automatic generation control , 2001, IEEE Trans. Control. Syst. Technol..

[28]  Farshid Keynia,et al.  Short-term load forecasting of power systems by combination of wavelet transform and neuro-evolutionary algorithm , 2009 .

[29]  Jing Zhao,et al.  Techniques of applying wavelet de-noising into a combined model for short-term load forecasting , 2014 .