A Hybrid One-step-ahead Time Series Model Based on GA-SVR and EMD for Forecasting Electricity Loads

Mu-Chiun  Shiu; Liang-Ying  Wei; Jing-Wei  Liu; Deng-Yang  Huang; Chien-Chih  Tu; Kuo-Hsiung  Liao

doi:10.6180/jase.2017.20.4.08

A Hybrid One-step-ahead Time Series Model Based on GA-SVR and EMD for Forecasting Electricity Loads

Computer Science and Information Engineering

Mu-Chiun Shiu¹, Liang-Ying Wei This email address is being protected from spambots. You need JavaScript enabled to view it.², Jing-Wei Liu³, Deng-Yang Huang², Chien-Chih Tu² and Kuo-Hsiung Liao²

¹Department of Information Management, National Yunlin University of Science and Technology, Yunlin, Taiwan 640, R.O.C.
²Department of Information Management, Yuanpei University of Medical Technology, Hsinchu, Taiwan 30015, R.O.C.
³Department of Sport Information and Communication, National Taiwan University of Sport, Taichung, Taiwan 404, R.O.C.

Received: August 24, 2016
Accepted: August 15, 2017
Publication Date: December 1, 2017

Download Citation: ||https://doi.org/10.6180/jase.2017.20.4.08

ABSTRACT

Economic growth increases the demand for electricity, and forecasting electricity loads is critical for providing cheaper electricity. Conventional time series methods have been applied to forecast electricity loads. However, traditional statistical methods, such as regression models, are unable to address nonlinear relationships, such as those of electricity loads. Moreover, most time series models which use electricity load data with many factors, such as climate conditions and region environments, involutedly would reduce the forecasting performance. To overcome these problems and improve the forecasting ability of time series models, this paper proposes a hybrid one-step-ahead time series model that is based on support vector regression (SVR), empirical mode decomposition (EMD), and a genetic algorithm (GA) to predict electricity loads. The experimental results were generated from 2 electricity load datasets from various countries, and the proposed model was compared with several models. Our findings indicate that the proposed model outperforms the other approaches in terms of mean absolute percentage error (MAPE).

Keywords: Electricity Load, Support Vector Regression, Empirical Mode Decomposition, Time Series, One-step-ahead Method, Genetic Algorithm

REFERENCES

[1] Cheng, C. H. and Wei, L. Y., “One Step-ahead ANFIS TimeSeries Model for Forecasting ElectricityLoads,” Optimization and Engineering, Vol. 11, pp. 303317 (2010). doi: 10.1007/s11081-009-9091-5
[2] Nowak, M.P.,Schultz,R.andWestphalen,A.,“AStochastic Integer Programming Model for Incorporating Day-ahead Trading of into Hydro-thermal Unit Commitment,”Optim Eng,Vol.6,pp.163176(2005).doi: 10.1007/s11081-005-6794-0
[3] Rnberg, R. N. and Misch, W. R., “A Two-stage Planning Model for Power Scheduling in a Hydro-thermal System under Uncertainty,” Optim Eng, Vol. 3, pp. 355378 (2002).
[4] Bunn, D. W. and Farmer, E. D., Comparative Models for Electrical Load Forecasting, New York: John Wiley and Sons (1985).
[5] Yang, Y., Chen, Y., Wang, Y., Li, C. and Li, L., “Modelling a Combined Method Based on ANFIS and Neural Network Improved by DE Algorithm: a Case Study for Short-term Electricity Demand Forecasting,” Applied Soft Computing, Vol. 49, pp. 663675 (2016). doi: 10.1016/j.asoc.2016.07.053
[6] Koprinska, I., Rana, M. and Agelidis, V. G., “Correlation and Instance Based Feature Selection for Electricity Load Forecasting,” Knowledge-Based Systems, Vol. 82, pp. 2940 (2015). doi: 10.1016/j.knosys. 2015.02.017
[7] Friedrich, L., Armstrong, P. and Afshari, A., “Midterm Forecasting of Urban Electricity Load to Isolate Air-conditioning Impact,” Energy and Buildings, Vol 80, pp. 7280 (2014). doi: 10.1016/j.enbuild.2014.05. 011
[8] Huang, S. J. and Shih, K. R., “Short-term Load Forecasting via ARMA Model Identification Including Non-Gaussian Process Considerations,” IEEE Trans Power Syst1, Vol. 18, No. 2, pp. 673679 (2003). doi: 10.1109/TPWRS.2003.811010
[9] Haida, T. and Muto, S., “Regression Based Peak Load Forecasting Using a Transformation Technique,” IEEE Trans Power Syst, Vol. 9, pp. 17881794 (1994). doi: 10.1109/59.331433
[10] Pino, R., Parreno, J., Gomez, A. and Priore, P., “Forecasting Next-day Price of Electricity in the Spanish Energy Market Using Artificial Neural Networks,” Engineering Applications of Artificial Intelligence, Vol. 21, pp. 5362 (2008). doi: 10.1016/j.engappai. 2007.02.001
[11] Fan,S.,Mao,C.andChen,L.,“PeakLoadForecasting Using the Self Organizing Map,” Advances in Neural Network, ISNN 2005. Springer, Part III, pp. 640647 (2005). doi: 10.1007/11427469_102
[12] Hippert, H. S., Pedreira, C. E. and Castro, R., “Neural Networks for Short-term Load Forecasting: a Review and Evaluation,” IEEE Trans Power Syst, Vol. 16, No. 1, pp. 4455 (2001). doi: 10.1109/59.910780
[13] Hsu, C. C. and Chen, C. Y., “Regional Load Forecasting in Taiwan- Applications of Artificial Neural Networks,” Energ Convers Manage, Vol. 44, No. 12, pp. 19411949 (2003). doi: 10.1016/S0196-8904(02) 00225-X
[14] Pai, P. F., “Hybrid Ellipsoidal Fuzzy Systems in Forecasting Regional Electricity Loads,” Energ Convers Manage, Vol. 47, No. 1516, pp. 22832289 (2006). doi: 10.1016/ j.enconman.2005.11.017
[15] Vapnik, V., The Nature of Statistical Learning Theory, New York: Springer-Verlag (1995). doi: 10.1007/ 978-1-4757-2440-0_2
[16] Pai, P. F. and Hong, W. C., “Forecasting Regional Electricity Load Based on Recurrent Support Vector Machines with Genetic Algorithms,” Electric Power Syst Res, Vol. 74, No. 3, pp. 417425 (2005). doi: 10.1016/ j.epsr.2005.01.006
[17] Ertugrul, Ö. F., “Forecasting Electricity Load by a Novel Recurrent Extreme Learning Machines Approach,” Electrical Power and Energy Systems, No. 78, pp. 429435 (2016). doi: 10.1016/j.ijepes.2015.12.006
[18] Ghasemi, A., Shayeghi, H., Moradzadeh, M. and Nooshyar, M., “A Novel Hybrid Algorithm for ElectricityPrice and Load Forecasting in SmartGrids with Demand-side Management,” Applied Energy, Vol. 177, pp. 4059 (2016). doi: 10.1016/j.apenergy.2016.05. 083
[19] Hassan, S., Khosravi, A., Jaafar, J. and Khanesar, M. A., “A Systematic Design of Interval Type-2 Fuzzy Logic System Using Extreme Learning Machine for Electricity Load Demand Forecasting,” International Journal of Electrical Power & Energy Systems, Vol. 82, pp. 110 (2016). doi: 10.1016/j.ijepes.2016.03.001
[20] Takeda, H., Tamura, Y. and Sato, S., “Using the Ensemble Kalman Filter for ElectricityLoad Forecasting and Analysis,” Energy, Vol. 104, pp. 184198 (2016). doi: 10.1016/j.energy.2016.03.070
[21] Clements, A. E. and Hurn, A. S., “Forecasting Dayahead Electricity Load Using a Multiple Equation Time Series Approach,” European Journal of Operational Research, Vol. 251, No. 2, pp. 522530 (2016). doi: 10.1016/j.ejor.2015.12.030
[22] Niu, D., Wang, Y. and Wu, D. D., “Power Load Forecasting Using Support Vector Machine and Ant Colony Optimization,” Expert Systems with Applications, Vol. 37, No. 3, pp. 25312539 (2010). doi: 10.1016/ j.eswa.2009.08.019
[23] Box, G. and Jenkins, G., Time Series Analysis: Forecasting and Control, Palgrave Macmillan, London, pp. 161215 (1976). doi: 10.1057/9781137291264_6
[24] Benaouda,D.andMurtagh,F.,“ElectricityLoadForecast Using Neural Network Trained from Wavelettransformed Data,” IEEE International Conference on Engineering of Intelligent Systems (2006). doi: 10. 1109/ICEIS.2006.1703163
[25] Enders, W., Applied Econometric Time Series, Wiley (2004). doi: 10.1198/tech.2004.s813
[26] Kim, K. J., “Financial Time Series Forecasting Using Support Vector Machines,” Neurocomputing, Vol. 55, No. 12, pp. 307319 (2003). doi: 10.1016/S09252312(03)00372-2
[27] Cao, L., “Support Vector Machines Experts for Time Series Forecasting,” Neurocomputing, Vol. 51, pp. 321 339(2003).doi:10.1016/S0925-2312(02)00577-5
[28] Chen, W. H. and Shih, J. Y., “AStudy of Taiwan’s Issuer Credit Rating SystemsUsing Support Vector Machines,” Expert Systems with Applications, Vol. 30, No. 3, pp. 427435 (2006). doi: 10.1016/j.eswa.2005. 10.003
[29] Drucker, H., Burges, C. J. C., Kaufman, L., Smola, A. and Vapnik, V., Support Vector Regression Machines, Advances in Neural Information Processing Systems, 9 Cambridge MA: MITpress (1997).
[30] Huang, W., Nakamori, Y. and Wang, S. Y., “Forecasting Stock Market Movement Direction with Support Vector Machine,” Computers and Operations Research, Vol. 32, No. 10, pp. 25132522 (2005). doi: 10.1016/ j.cor.2004.03.016
[31] Pai, P. F. and Lin, C. S., “A Hybrid ARIMA and Support Vector Machines Model in Stock Price Forecasting,” Omega, Vol. 33, No. 6, pp. 497505 (2005). doi: 10. 1016/j.omega.2004.07.024
[32] Wang,J.,Houb, R.,Wang,C.andShen,L.,“Improved v -Support Vector Regression Model Based on Variable Selection and Brain Storm Optimization for Stock Price Forecasting,” Applied Soft Computing, Vol. 49, pp. 164–178 (2016). doi: 10.1016/j.asoc.2016.07.024
[33] Wei, L. Y., “A Hybrid Model Based on ANFIS and Adaptive Expectation Genetic Algorithm to Forecast TAIEX,” Economic Modelling, Vol. 33, pp. 893899 (2013). doi: 10.1016/j.econmod.2013.06.009
[34] Holland, J. H., Adaptation in Nature and Artificial Systems, University of Michigan Press (1975). doi: 10.1145/1216504.1216510
[35] Chang, F. J., Chiang, Y. M. and Chang, L. C., “Multistep-ahead Neural Networks for Flood Forecasting,” Hydrological Sciences-Journal-des Sciences Hydrologiques, Vol. 52, pp. 114130 (2007). doi: 10.1623/ hysj.52.1.114
[36] Huang, N. E., Shen, Z., Long, S. R., Wu, M. C., Shih, H.H.andZheng, Q.,“The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Nonstationary Time Series Analysis,” in: Proceedings of the Royal Society of London Seriesa-Mathematical Physical and Engineering Sciences,SeriesA,Vol.454, pp. 903995 (1998). doi: 10.1098/rspa.1998.0193
[37] Vincent, H. T., Hu, S.-L. J. and Hou, Z., “Damage Detection Using Empirical Mode Decomposition Method and a Comparison with Wavelet Analysis,” in: Proceedings of the Second International Workshop on Structural Health Monitoring, Stanford, pp. 891900 (1999).
[38] Huang, N. E., Shen, Z. and Long, S. R., “ANew View of Nonlinear Water Waves: the Hilbert Spectrum,” Annu. Rev. Fluid Mech., Vol. 31 pp. 417457 (1999). doi: 10.1146/annurev.fluid.31.1.417
[39] Yu, D. J., Cheng, J. S. and Yang, Y., “Application of EMD Method and Hilbert Spectrum to the Fault Diagnosis of Roller Bearings,” Mech. Syst. Signal Process, Vol. 19, No. 2, pp. 259270 (2005). doi: 10.1016/ S0888-3270(03)00099-2
[40] Kazema, A., Sharifia, E., Hussainb, F. K., Saberic, M. and Hussaind, O. K., “Support VectorRegression with Chaos-basedFireflyAlgorithmforStockMarketPrice Forecasting,” Applied Soft Computing, Vol. 13, No. 2, pp. 947958 (2013). doi: 10.1016/j.asoc.2012.09.024
[41] Allen, F. and Karalainen, R., “Using Genetic Algorithms to Find Technical Trading Rules,” J Financ Econ, Vol. 51, pp. 245271 (1999).
[42] Kim, M. J., Min, S. H. and Han, I., “An Evolutionary Approach to the Combination of Multiple Classifiers to PredictaStock PriceIndex,” Expert Syst. Appl, Vol. 31, No. 2, pp. 241247 (2006). doi: 10.1016/j.eswa. 2005.09.020
[43] Cheng, C. H., Chen, T. L. and Wei, L. Y., “A Hybrid Model Based on Rough Sets Theory and Genetic AlgorithmsforStockPriceForecasting,”InformSciences, Vol. 180, No. 9, pp. 16101629 (2010). doi: 10.1016/ j.ins.2010.01.014
[44] E. Goldberg, D., Genetic Algorithms in Search, Optimization and Mach Learn, Reading, MA: Addison Wesley (1989).
[45] Taylor, J. W. and Buizza, R., “Using Weather Ensemble Predictions in Electricity Demand Forecasting,” Int J Forecasting, Vol. 19, pp. 5770 (2003). doi: 10. 1016/S0169-2070(01)00123-6
[46] Poland Dataset, Electricityload values of Poland from 1990’s. <http://research.ics.aalto.fi/eiml/datasets.shtml> (accessed 3.01.2017).
[47] Yaslan, Y. and Bican, B., “Empirical Mode Decomposition Based Denoising Method with Support Vector Regression for Time Series Prediction: a Case Study for Electricity Load Forecasting,” Measurement, Vol. 103, pp. 5261 (2017). doi: 10.1016/j.measurement. 2017.02.007