Applied-Information Technology in Power System with Short-Term Load Forecasting Based on SPSS and BP Neural Network

Abstract:

In the study of power load forecasting, the factors influencing power load have data redundancy and data nonlinearity. The traditional load forecasting methods can’t eliminate redundant or nonlinear law between data, which result in reduced accuracy. In order to improve the predictive accuracy of power load, a prediction model based on BP neural network and SPSS (SPSS-BP) is established. The paper first analyzes the correlation and principal component of influence factors of electric power load, which eliminates the redundancy between various factors, accelerates the speed of BP neural network forecasting and improves predictive accuracy; then model the processed data and forecast through the BP neural network model. One-month weather data and load data of Yichang city have been confirmatory tested and analyzed through application of SPSS-BP model. The results show that SPSS-BP model significantly improves the accuracy, verify the feasibility and effectiveness of the model.

• Periodical: Advanced Materials Research (Volume 1022)
• Pages: 241-244
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1022.241
• Online since: August 2014
• Authors: Jian Ping Chen*, Chang Hao Xia, Zhi Peng Tian
• Keywords: BP Neural Network (BPNN), Load Forecasting, Principal Component Analysis (PCA), SPSS

References

[1] Ruiming Fang. SVM Theory and Application Analysis. Beijing: china power press, (2004).

[2] Mi H, Wenzhang Zhang. The utility of modern statistical analysis methods and the application of SPSS. Beijing: contemporary china publishing house, (2000).

[3] Zhao X, Wen X j, Shao H H. Combination method of principal component analysis and support vector Machine for on-line process monitoring and fault diagnosis. Journal of Donghua University (EnglishEdition), 2006, (01).

[4] Shuangfeng Ye. Application and consideration about principal component analysis. Application of Statistics and Management, 2001, 20(2): 52−56.

[5] Deduth H, Beatle M. Neural network toolbox for use with MATLAB. MA: The Math Works Inc, 2001: 118−130.

[6] Xing WEN. Neural network simulation and application based on MATLAB. Beijing: Science Press, 2003: 278−283.

[7] Wang H Q, Song H Z, Li P. improved PCA with application to process monitoring and fault diagnosis. Journal of Chemical Industry and Engineering (China), 2001, 52(6), 471-475.

[8] Liu Z L, Liu G Z, Liu J. Adaptive tracking controller using BP neural networks for a class of nonlinear systems. Journal of Systems Engineering and Electronics (English Edition). 2004(04).

[9] Jianying Yu, HE Xu-hong. Data statistical analysis and application of SPSS. Beijing: People's Posts and Telecommunications Press, 2003: 291-310.

[10] Zixiang Xu, Deyun ZHOU. Fuzzy neural network based on principal component. Computer Engineering and Applications, 2006, 42(5): 34−36.