A Novel Electricity Sales Forecasting Method Based on Clustering, Regression and Time-series Analysis

The forecasting of monthly electricity sales is a fundamental job for the marketing department of State Grid Corporation of China. In this paper, we propose a novel electricity sales forecasting method. Firstly, with the visual clustering algorithm, 27 provincial electric power companies of State Grid Corporation of China are clustered into different groups according to their historical monthly electricity sales curve; secondly, as to different groups, the monthly electricity sales are predicted by specific regression or time-series algorithms. Experiments show that the proposed approach has very high prediction accuracy rate in terms of forecasting the the electricity sales of the next 12 months for the power companies of State Grid Corporation of China.

[1]  Zongben Xu,et al.  A heuristic hierarchical clustering based on multiple similarity measurements , 2013, Pattern Recognit. Lett..

[2]  V. Vapnik,et al.  Bounds on Error Expectation for Support Vector Machines , 2000, Neural Computation.

[3]  Sanjay Jharkharia,et al.  Applicability of ARIMA Models in Wholesale Vegetable Market: An Investigation , 2013, Int. J. Inf. Syst. Supply Chain Manag..

[4]  Nataliya Kalashnykova,et al.  Liberalization of Mexican Electricity Market: Various Modeling Approaches with Numerical Simulation Results , 2008, 2008 Eighth International Conference on Intelligent Systems Design and Applications.

[5]  Zongben Xu,et al.  Hierarchical clustering driven by cognitive features , 2013, Science China Information Sciences.

[6]  Kazuhiko Kakamu,et al.  Forecasting electricity demand in Japan: A Bayesian spatial autoregressive ARMA approach , 2010, Comput. Stat. Data Anal..

[7]  Sayan Mukherjee,et al.  Choosing Multiple Parameters for Support Vector Machines , 2002, Machine Learning.

[8]  Hamidreza Zareipour,et al.  Electricity Price and Demand Forecasting in Smart Grids , 2012, IEEE Transactions on Smart Grid.

[9]  Bernard Widrow,et al.  Neural networks: applications in industry, business and science , 1994, CACM.

[10]  Shuang Liu,et al.  The study of long-term electricity load forecasting based on improved grey prediction model , 2013, 2013 International Conference on Machine Learning and Cybernetics.

[11]  Zongben Xu,et al.  $L_{1/2}$ Regularization: A Thresholding Representation Theory and a Fast Solver , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[12]  Matthew Scotch,et al.  Comparison of ARIMA and Random Forest time series models for prediction of avian influenza H5N1 outbreaks , 2014, BMC Bioinformatics.

[13]  Kazuhiro Ozawa,et al.  Electricity Demand and Price Analysis in California Using Possibility Regression Model , 2003, J. Adv. Comput. Intell. Intell. Informatics.

[14]  John V. Ringwood,et al.  Forecasting Electricity Demand on Short, Medium and Long Time Scales Using Neural Networks , 2001, J. Intell. Robotic Syst..

[15]  David E. Rumelhart,et al.  Product Units: A Computationally Powerful and Biologically Plausible Extension to Backpropagation Networks , 1989, Neural Computation.

[16]  Zongben Xu,et al.  L1/2 regularization , 2010, Science China Information Sciences.

[17]  James W. Taylor,et al.  Triple seasonal methods for short-term electricity demand forecasting , 2010, Eur. J. Oper. Res..