Daily Peak Load Forecasting Based on Fast K-medoids Clustering, GARCH Error Correction and SVM Model

Zongyun Song; Dongxiao  Niu; Xinli Xiao; Han Wu

doi:10.6180/jase.2016.19.3.02

Daily Peak Load Forecasting Based on Fast K-medoids Clustering, GARCH Error Correction and SVM Model

Computer Science and Information Engineering

Zongyun Song This email address is being protected from spambots. You need JavaScript enabled to view it.¹, Dongxiao Niu¹, Xinli Xiao¹ and Han Wu¹

¹School of Economic and Management, North China Electric Power University, Beijing 102206, P.R.China

Received: January 18, 2015
Accepted: January 21, 2016
Publication Date: September 1, 2016

Download Citation: ||https://doi.org/10.6180/jase.2016.19.3.02

ABSTRACT

Safe and economicoperation of power systemis based on load forecasting, and how to increase forecasting accuracy is the premise of power dispatching and economic analysis. Present paper establishes SVM (support vector machine) forecasting model based on fast K-medoids clustering algorithm and data accumulative pre-processing. FKM (fast K-medoids clustering algorithm) is applied to extract similar days by dividing all samples into k clusters, and respective forecasting of k clusters can realize the forecasting of a whole object. Before inputting the data into SVM system, the original data is preprocessed by accumulation to weaken the irregularity disturbance and strengthen sequence regularity. Due to existing unexplained component in forecasting error, GARCH (generalized autoregressive conditional heteroskedasticity) model is employed to forecast the error with non-white noise. According to its results, error correction is applied to the forecasted daily peak load. The forecasting effect of the proposed model is compared with other models in the given example, which verifies that SVM model based on fast K-medoids clustering algorithm and GARCH model has the characteristic of effectiveness, superiority and universality. The accuracy of daily peak load forecasting is enhanced significantly.

Keywords: Daily Peak Load Forecasting, FKM, SVM, GARCH, Error Correction

REFERENCES

[1] Kavousi-Fard, A., “A New Fuzzy-based Feature Selection and Hybrid TLA-ANN Modelling for Short term Load Forecasting,” Journal of Experimental & Theoretical Artificial Intelligence, Vol. 25, No. 4, pp. 543557(2013).doi:10.1080/0952813X.2013.782350
[2] Akbari-Zadeh, M. R., “A Hybrid Method Based on Wavelet, ANN and ARIMAModel for Short-termLoad Forecasting,” Journal of Experimental & Theoretical Artificial Intelligence, Vol. 26, No. 2. pp. 167182 (2014). doi: 10.1080/0952813X.2013.813976
[3] Kavousi-Fard, A., Khosravi, A. and Nahavandi, S., “A New Fuzzy-based Combined Prediction Interval for Wind Power Forecasting,” IEEE Transactions on Power Systems, Vol. 31, No. 1, pp. 1826 (2015). doi: 10. 1109/TPWRS.2015.2393880
[4] Xiao, Z., Ye, S. J., Zhong, B. and Sun, C. X., “BPNeural Network with Rough Set for Short Term Load Forecasting,” Expert Systems with Applications, Vol. 36, No. 1, pp. 273279 (2009). doi: 10.1016/j.eswa. 2007.09.031
[5] Feng, Y. and Xu, X., “A Short-term Load Forecasting Model of Natural Gas Based on Optimized Genetic Algorithm and Improved BP Neural Network,” Apply Energy, Vol. 134, No. 134, pp. 102113 (2014). doi: 10.1016/j.apenergy.2014.07.104
[6] Hwang, H., “Daily Electric Load Forecasting Based on RBF Neural Network Models,” International Journal of Fuzzy Logic & Intelligent Systems, Vol. 13, No. 1, pp. 3949 (2013). doi: 10.5391/IJFIS.2013.13.1.39
[7] Yao, Y., Lian, Z., Hou, Z. and Liu, W., “An Innovative Air-conditioning Load Forecasting Model Based on RBF Neural Network and Combined Residual Error Correction,” International Journal of Refrigeration, Vol. 29, No. 4, pp. 528538 (2006). doi: 10.1016/j. ijrefrig.2005.10.008
[8] Vapnik, V., Golowich, S. E. and Smola, A., “Support Vector Method for Function Approximation, Regression Estimation and Signal Processing,” Advances in Neural Information Processing Systems, Vol. 9, pp. 281287 (1997).
[9] Lee, C. M. and Ko, C. N., “Time Series Prediction Using RBF Neural Networks with a Nonlinear Time varying Evolution PSO Algorithm,” Neurocomputing, Vol. 73, No. 1, pp. 449460 (2009). doi: 10.1016/j. neucom.2009.07.005
[10] Avci, E., “Selecting of the Optimal Feature Subset and Kernel Parameters in Digital Modulation Classification by Using Hybrid Genetic Algorithmsupport Vector Machines: HGASVM,” Expert Systems with Applications, Vol. 36, No. 2, pp. 13911402 (2009). doi: 10. 1016/j.eswa.2007.11.014
[11] Chen, S. X., Yang, Z., Zhu, J. and Li, F. W., “Network Security Situation Prediction Method Based on PSOSVM,” Application Research of Computers, Vol. 32, No. 6, pp. 17781781(2015).
[12] MacQueen, J. B., “Some Methods for Classification and Analysis of Multivariate Observations,” The 5th Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1, pp. 281297 (1967).
[13] Chen, X. Q., Peng, H. and Hu, J. S., “K-medoids Substitution Clustering Method and a New Clustering Validity Index Method,” The 6th World Congress on Intelligent Control and Automation, Vol. 2, pp. 5896 5900 (2006). doi: 10.1109/WCICA.2006.1714209
[14] Park, H. S. and Jun, C. H., “A Simple and Fast Algorithm for K-medoids Clustering,” Expert Systems with Applications, Vol. 36, No. 2, pp. 33363341 (2009). doi: 10.1016/j.eswa.2008.01.039
[15] Li, X. Y. and Fu, Y., “Improved K-medoids Algorithm,” Journal of Chengdu University of Information Technology, Vol. 21, No. 4, pp. 532534 (2006).
[16] Xie, J. Y. and Gao, R., “K-medoids Clustering Algorithmswith Optimized Initial Seeds by Variance,” Journal of Frontiers of Computer Science and Technology, No. 8, pp. 973984 (2015).
[17] Liu,S.D., He,E.D., Liu,L.and Xiao,S.Y.,“Forecasting Electricity Futures Price with Cointegration Model,” Journal of Electric Power Science and Technology, Vol. 23, No. 2, pp. 4650 (2008).
[18] Wang, P., Tai, N. L., Wang, B., Huo, H. Q., Ye, J., Li, L., Zhu, J. D. and Qi, J. B., “A Short-term Load Forecasting Correction Method Considering Climatic Factors,” Automation of Electric Power Systems, Vol. 32, No. 13, pp. 9296 (2008).
[19] Chen, G. D., Yao, J. G., Qian, W. H. and Long, L. B., “Research on Load Forecasting Based on Predicted Error Amendment,” Modern Electric Power, Vol. 24, No. 3, pp. 1115 (2007).
[20] Li, M. J., “Forecasting and Evaluation on Energy Efficiency of China by a Hybrid Forecast Method,” Energy Procedia, Vol. 75, pp. 27242730 (2015). doi: 10.1016/j.egypro.2015.07.703
[21] Liu, D., Niu, D. X., Xing, M., et al., “Day-ahead Price Forecast with Genetic Algorithm Optimized Support Vector Machines Based on GARCH Error Calibration,” Automation of Electric Power Systems, Vol. 31, No. 11, pp. 3134 (2007).
[22] Sang, X. L., Li, Z., Xiao, H. J., et al., “CPI Combination Forecasting Model Based on ARIMA-GARCH and Parabola Model,” Statistics and Decision, No. 15, pp. 2325 (2015).
[23] Zhang, C., “Stock Price Forecast with ARMAGARCH Based on Error Correction,” Journal of Nanjing University of Aeronautics and Astronautics (Social Sciences), Vol. 16, No. 3, pp. 4348 (2014).
[24] Viet, V. Q., Thang, H. M. and Choi, D., “Personalization in Mobile Activity Recognition System Using KMedoids Clustering Algorithm,” International Journal of Distributed Sensor Networks, Vol. 2013, No. 9, pp. 797800 (2013). doi: 10.1155/2013/315841
[25] Vapnik, V., The Nature of Statistical Learning Theory, 1st ed., Springer New York, New York, pp. 988999 (1995). doi: 10.1007/978-1-4757-2440-0
[26] Bollerslev, T., “Generalized Autoregressive Conditional Heteroskedasticity,” Journal of Economics, Vol. 31, pp. 307327 (1986). doi: 10.1016/0304-4076(86)90063-1
[27] Zhu, S. M., Yang, M. and Han, X. S., “Short-Term Generation Forecast of Wind Farm Using SVMGARCH Approach,” 2012 IEEE International Conference on Power System Technology, Auckland, U.S.A., Oct. 30Nov. 2, pp. 16 (2012). doi: 10.1109/Power Con.2012.6401309
[28] Lin, Y. Y., Yi, T., Chen, B., et al., “Short-term Load Forecasting Considering Meteorological Factors and Date Type,” Computer Simulation, Vol. 31, No. 3, pp. 109112 (2014).
[29] Li, C. B., Li, X. H., Zhao, R., et al., “A Novel Algorithm of Selecting Similar Days for Short-term Power Load Forecasting,” Automation of Electric Power Systems, Vol. 32, No.9, pp. 6973 (2008).
[30] Li, L., Wei, J., Li, C. B., et al., “Prediction of Load Model Based on Artificial Neural Network,” Transactions of China Electrotechnical Society, Vol. 30, No. 8, pp. 225230 (2015).