AUTHORS: Jianfeng Li, Dongxiao Niu, Ming Wu, Yongli Wang, Meng Li, Mingyue Yong

Download as PDF

ABSTRACT: With the development and improvement of the electricity market, the role of power load forecasting in competitive market is being recognized. It is very important to ensure the security and stability of electric network by accurate load forecasting. Meanwhile, the reform of the electricity market leads to a result that it is far more difficult for the match between power supply-side and demand-side, which brings new challenges to the power load forecasting industry. According to the chaotic and non-linear characters of power load data, and to prove the effectiveness and accuracy, the model of AD-SVM (average displacement-support vector machines) based on Lyapunov exponents and wave approximation function optimization was established. Firstly, the optimal time delay and embedding dimension sequence of time series are determined by average displacement method. Secondly the method of maximum Lyapunov exponents is devoted to selecting the appropriate embedding dimension and time delay, which proves the effectiveness of the accuracy invariants of phase-space; and then considering the volatility autocorrelation characteristics, we determine learning parameters based on the exploiting fluctuation approximation function and embedding dimension, which can greatly assist us with the prediction of future market; meanwhile, article applies KNN method to estimate noise standard deviation so as to determine the parameter selection. Finally, support vector machine is utilized for establishing the model of power load forecasting. In addition, for existing generalized autoregressive heteroscedasticity of phenomenon of LSSVM prediction error, In this paper, the prediction error is modified by GARCH model on the basis of the original and the prediction error of the auto regressive process is eliminated. Therefore, it can further enhance prediction accuracy. In order to prove the rationality of chosen dimension and the effectiveness of the model, any other random dimensions were selected to compare with the calculated dimension while BP algorithm was used to compare with the result of SVM. The results show that the model is effective and more accurate in the forecasting of short-term power load

KEYWORDS: Power load forecasting; Support vector machines; Lyapunov exponents; average displacement; Wave approximate function; embedding dimension; GARCH error correction

REFERENCES:

[1] Dongxiao Niu, Shuhua Cao, and Yue Zhao, Technology and Application of Power Load Forecasting, Beijing: China Power Press, 1998.

[2] Wang Hui-zhong, Zhou Jia, LiuKe.Summary of Research on the Short-term Load Forecasting Method of the Electric Power System. Electrical Automation,37(1)(2015)36-38

[3] Zhou Chao. Summarization on Load Forecasting Method of Electrical Power System. Power Supply Corporation,(2012)32-39

[4] Wang Ji-cheng. Research on Chaos Forecast Methods of Load of Electrical Power System. Applied Technology, 2013.pp.75-76

[5] Liu xi-zhe. Study of the Distributed Power System Load Forecasting Based on Choas Theory (D). Changsha University, 2011.

[6] Zhao D F, Wang M, Zhang J S, et al. A support vector machine approach for short term load forecasting

[J]. Proceedings of the Csee, 22(4)(2002)26-29

[7] Yuan-Cheng L I, Fang T J, Er-Keng Y U. Study of support vector machines for short-term load forecasting. Proceedings of the Csee. Proceedings of the Csee, 2003.pp.654-659.

[8] Wang Bao-cheng. A Predict on Power Load in A Short Term Based on SVR. Times Agricultural Machinery,42(5)(2015)51-52

[9] Qiu Cun-yong, Xiao Jian. Power System Short-Term Load Forecasting Based on Support Vector Regression. Computer Simulation,30(11) (2013)62-64

[10] Wang Y J, Dian-Wen L I. Research on the Short-Term Load Forecast Based on the Improved SVM. Electrical Measurement & Instrumentation, 51(18)(2014)1-4

[11] Niu D, Lv J L, Zhang X. The Improved Markov Error Correcting Method in Gray SVM for Power Load Forecasting// International Symposium on Intelligent Information Technology Application Workshops. IEEE Computer Society, 2008.pp.793-796.

[12] Wang De-yi,Yang Zhuo, Yang Guo-qing. Short-Term Load Forecasting Based on Chaotic Characteristic of Loads and Least Squares Support Vector Machines. Power System Technology, 32(7)(2008)66-69.

[13] Grant, Jason Lee. Short-term peak demand forecasting using an artificial neural network with controlled peak demand through intelligent electrical loading. University of Miami, (2014)632-635.

[14] Quan H, Srinivasan D, Khosravi A. Short-term load and wind power forecasting using neural network-based prediction intervals. IEEE Transactions on Neural Networks & Learning Systems, 25(2)(2014)303-315.

[15] Wang J, Jia R, Zhao W, et al. Application of the largest Lyapunov exponent and non-linear fractal extrapolation algorithm to short-term load forecasting. Chaos Solitons & Fractals, 45(s 9–10) (2012)1277-1287.

[16] Liu kui. Research on short-term load forecasting method based on chaos and wavelet neural network. Southwest Jiaotong University Master Degree Thesis, 2012.

[17] Wang J, Zhou J, Bing P. A hybrid neural genetic method for load forecasting based on phase space reconstruction. Kybernetes the International Journal of Systems & Cybernetics, 39(8)(2010)1291-1297.

[18] Wang J, Jia R, Zhao W, et al. Application of the largest Lyapunov exponent and non-linear fractal extrapolation algorithm to short-term load forecasting. Chaos Solitons & Fractals, 45(s 9–10) (2012)1277-1287.

[19] Li Dongdong, Qin Zishan. Short-term Load Forecasting for Microgrid Based on Method of Chaotic Time Series. Proceedings of the CSU-EPSA, 27(5)(2015)14-17.

[20] Li Lan-you,Du Jie, Shao Ding-hong, LU Jin-gui. Larges Lyapunov exponent of short-term electric load forecasting model simulation. Computer Engineering and Design, 28(12)(2007)5934-5938.

[21] Fan Z, Qin Z, Hong T, et al. The studying of combined power-load forecasting by error evaluation standard based on RBF network and SVM method// Control and Decision Conference, CCDC '09. Chinese. 2009.pp.4016-4019.

[22] Wu J, Niu D. Short-Term Power Load Forecasting Using Least Squares Support Vector Machines(LS-SVM)// Computer Science and Engineering,2009. WCSE '09.Second International Workshop on. IEEE, 2009.pp.246-250.

[23] Zhang Q. Application of Support Vector Machine and Fuzzy Rules Method for Power Load Forecasting// International Conference on Computer Engineering & Applications. IEEE Computer Society, 2010.pp.542-545.

[24] Xu Y, Du P. Researching about Short-Term Power Load Forecasting Based on Improved BP ANN Algorithm// icise. IEEE Computer Society,2009.pp.4094-4097.

[25] Sun W, Zhao W. Mid-long term power load forecasting based on MG-CACO and SVM method// Future Computer and Communication (ICFCC), 2010 2nd International Conference on. IEEE, 2010.pp.V1-118-V1-121.

[26] Quan Wen, Yangchuan Zhang, and Shijie Chen, “The analysis approach based on chaotic time series for load forecasting,” Power System Technology, 2001.pp.13-16.

[27] Zhishan Liang, Liming Wang, and Dapeng Fu, “Short-term power load forecasting based on lyapunov exponents,” Proceeding of the CSEE,1998.pp.368-472.

[28] Francis E. H. Tay, Lijuan Cao. Application of Support Vector Machines in Financial Time Series Forecasting. The International Journal of Management Science, 2001.pp.309-317.

[29] Kodba Stane, Perc Matjaz, Marhl Marko. Detecting chaos from a time series. European Journal of Physics 26(1) (2005)205-215.

[30] Wolf A, Swift J B and Swinney H L. Determining lyapunov exponents from a time series. Physics D, 1985.pp.285-317.

[31] Liu Qiming, Jiamalihan·Kumashi, Hua Dong, Li Pengfei. Load Forecasting Based on Support Vector Machine Analysis. Electric technology, 2013.pp.34-36

[32] Zhang Xuegong. Nature of Statistical Learning Theory. Beijing tsinghua university press , 2000.

[33] Zhang L, Liu X S, Yin H J. Application of support vector machines based on time sequence in power system load forecasting. Power System Technology, 28(19)(2004)38-41.

[34] .Li Xin. Study on power load forecasting model based on phase space reconstruction and SVM. Electrical Measurement ＆ Instrumentation, 51(24)(2014)6-8

[35] Dongning Tan and Donghan Tan, “Small-sample machine learning theory-statistical learning theory,” Journal of Nanjing University of Science and Technology,25(1)(2001)108-112.

[36] Zhou You, Xiang Jinglin. A Quickly Converging and Numerically Stable Method of Calculating Finite Time Lyapunov Exponents (FTLEs) of Multidimensional System with Degenerate Spectrum. Journal of Northwestern Poly technical University. 23(2)(2005)240-243.

[37] Huo Ming. Research on Parameters Optimization of SVM Model for Short-term Loading Forecasting. Hunan University, 2009.