Conventional regression versus artificial neural network in short-term load forecasting

In order to short-term load forecasting (STLF), two different seasonal artificial neural networks (ANNs) are designed and compared with conventional regression. Furthermore designed ANNs are compared with each other in terms of model complexity, robustness and forecasting accurate to make more accurate short-term load forecasting in electricity market of Iran. The first model is a daily forecasting model which is used for forecasting the hourly load of the next day, and the second model is comprised of 24 sub-networks which are used for forecasting the hourly load of the next day. In fact, the second model is partitioning the first model. Time, temperature, and historical loads are taken as inputs. Results show a good conformity between actual data and ANNs outcome. In comparison between ANNs results and regression's, even thought in some cases regression shows less MAPE, the total MAPE of ANN is proved to be less than regression's. Moreover, it is founded that the first designed model consisting of single ANN is more appropriate than the second model consisting of 24 ANNs.

[1]  S. A. Villalba,et al.  Hybrid demand model for load estimation and short term load forecasting in distribution electric systems , 2000 .

[2]  Junaid Khan,et al.  Fuzzy neural very-short-term load forecasting based on chaotic dynamics reconstruction , 2006 .

[3]  Ali Azadeh,et al.  A simulated-based neural network algorithm for forecasting electrical energy consumption in Iran , 2008 .

[4]  T. Senjyu,et al.  Neural networks approach to forecast several hour ahead electricity prices and loads in deregulated market , 2006 .

[5]  Moon-Hee Park,et al.  Short-term Load Forecasting Using Artificial Neural Network , 1992 .

[6]  Mehmet Kurban,et al.  A New Approach for the Short-Term Load Forecasting with Autoregressive and Artificial Neural Network Models , 2007 .

[7]  A. Kazemi,et al.  Day-ahead price forecasting in restructured power systems using artificial neural networks , 2008 .

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

[9]  R. Buizza,et al.  Using weather ensemble predictions in electricity demand forecasting , 2003 .

[10]  Desheng Dash Wu,et al.  Power load forecasting using support vector machine and ant colony optimization , 2010, Expert Syst. Appl..

[11]  Shyh-Jier Huang,et al.  Short-term load forecasting via ARMA model identification including non-Gaussian process considerations , 2003 .

[12]  Rey-Chue Hwang,et al.  A new artificial intelligent peak power load forecaster based on non-fixed neural networks , 2002 .

[13]  Maarouf Saad,et al.  An efficient approach for short term load forecasting using artificial neural networks , 2006 .

[14]  K. S. Swarup,et al.  Effect of temperature on short term load forecasting using an integrated ANN , 2004 .

[15]  Bo Zhong,et al.  BP neural network with rough set for short term load forecasting , 2009, Expert Syst. Appl..

[16]  S. Vemuri,et al.  Neural network based short term load forecasting , 1993 .

[17]  Omar Badran,et al.  A fuzzy inference model for short-term load forecasting , 2009 .

[18]  Ching-Hsue Cheng,et al.  One step-ahead ANFIS time series model for forecasting electricity loads , 2010 .

[19]  Rey-Chue Hwang,et al.  An adaptive modular artificial neural network hourly load forecaster and its implementation at electric utilities , 1995 .

[20]  Jae Hong Park,et al.  Composite modeling for adaptive short-term load forecasting , 1991 .

[21]  Tomonobu Senjyu,et al.  Next day load curve forecasting using recurrent neural network structure , 2004 .

[22]  Jingfei Yang,et al.  Short-term load forecasting with increment regression tree , 2006 .