Short Term Load Forecasting Based on Fuzzy Clustering

Abstract:

This paper presents selected similar days of short-term power load forecasting model based on fuzzy clustering, the method first meteorological factors subdivided into temperature, barometric pressure, wind speed, rain, etc., and then type the week, date, type of day together constitute similar factors, fuzzy coefficient feature mapping table through fuzzy rules, not only to achieve quantitative impact factors and facilitate real-time to add a new rule. On this basis, the use of fuzzy clustering method for classification, based on the level of clustering similar day is selected to reduce the number of samples to accelerate the selected speed. The model takes into account the full impact of weather and other factors on load forecasting, further weakening the load of randomness. Simulation results show that the method has higher prediction accuracy.

• Periodical: Applied Mechanics and Materials (Volumes 672-674)
• Pages: 1413-1420
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.672-674.1413
• Online since: October 2014
• Authors: Yi Jun Wang, Cheng Lu, Dian Wen Li
• Keywords: Fuzzy Clustering, Short-Term Load Forecasting, Similar Day

References

[1] Tai Nengling, Hou Zhijian．Application of wavelet fuzzy neural network in short term load forecasting of power system[J]．Proceedings of the CSEE，2004，24(1)：24-29.

[2] Yao Lixiao, Yao Jinxiong, Li Baoqing etal. Competitive category-based neural network short term load forecasting[J]． Power System Technology，2004，28( 10)：45-48.

[3] Li Yuancheng, Fang Tingjian, Yu Erqiang．Research on an SVM method of short term load forecasting[J]．Proceedings of the CSEE，2003，23(6)：55-59.

[4] Xie Jiecheng, Zhang Dali, Xu Wenli. An review on wavelet image denoising［J］. Journal of Image and Graphics，2002，7（3）：209-217.

[5] Kang Chongqing, Xia Qing, Liu Mei. Load forecasting of power system［M］. Beijing：China Electric Power Press，(2007).

[6] Zhang Xiaoxing, Cheng Qiyun, Zhou Quan etal. Chaotic neural network short term load forecasting with an adjustable excitation function[J]. High Voltage Technology, 2005, 31(7): 51-54.

[7] Li Tianyun, Zhao Yan, Li Nan. EMD-based Hilbert transform applies on transient signal analysis[J]. Automation of Electric Systems, 2004, 29(4): 49-51.

[8] Ding Jianyong, Liu Yun. Neural network short term load forecasting based on load feature extraction[J]. High Voltage Technology, 2004, 30(12): 47-49.

[9] Yuan Yuan. Power load features analysis and middle-short term load forecasting research method[D]. Shanghai：Shanghai Jiao Tong University，(2008).

[10] Chapelle O，Vapnik V．Choosing multiple parameters for support vector machines[R]．New York：AT＆T Research Labs，2001.

[11] Niu Dongxiao, Cao Shuhua, Zhao Lei etal. Power load forecasting technology and its application[M]. Beijing: China Electric Power Press, (2009).

[12] Ma Jingbo, Yang Honggeng. Application of self-adaptive Kalman filtering in short term load forecasting of power system[J]. Power System Technology, 2005, 29(1): 75-79.

[13] Zhang Junfang, Wu Yiang, Wu Junji. Applied research on load forecasting based on grey theory[J]. Automation of Electric Systems, 2004, 24(5): 24-27.

[14] Xu Zhiyuan, Ren Lina, Feng Ruicheng. A load forecasting model of power system in Gansu based on Elman neural network[J]. Modern Electric Power, 2007, 24(2): 26-29.

[15] Song Chao, Huang Minxiang, Ye Jianbin? Application of the wavelet analysis method in short term load forecasting of power system[J]. Proceedings of the CSU-EPSA , 2002, 14(3): 8-12.