Education, Science, Technology, Innovation and Life
Open Access
Sign In

Ultra Short Term Load Forecasting Based on Optimized Weight Cubature Kalman Filter and Support Vector Machine Combination Model

Download as PDF

DOI: 10.23977/jaip.2020.040206 | Downloads: 4 | Views: 174

Author(s)

Bei Huang 1,2, Weidong Kang 1, Hao Gu 1, Boyang Zhou 1,2, Shi Chen 1,2

Affiliation(s)

1 Anhui Nanrui Jiyuan Power Grid Technology Co., Ltd., China
2 Anhui Huahong Information Technology Co., Ltd., China

Corresponding Author

Hao Gu

ABSTRACT

In this paper, a combined ultra short term load forecasting model is proposed to solve the problem of less feature dimension and unclear relationship in ultra short term load forecasting for industrial power users. The model combines the cubature Kalman filter (CKF) prediction method which is better in nonlinear dynamic system and the least squares support vector machine (LS-SVM) prediction method which is better in small-scale data prediction. It combines the advantages of the two algorithms by using the combination of grey neural network, and avoids a single algorithm falling into local optimum. It combines horizontal prediction with vertical prediction Finally, the combination model is better than the single prediction.

KEYWORDS

Cubature kalman filter, Least squares support vector machine, Grey neural network, Combinatorial model

CITE THIS PAPER

Bei Huang, Weidong Kang, Hao Gu, Boyang Zhou, Shi Chen. Ultra Short Term Load Forecasting Based on Optimized Weight Cubature Kalman Filter and Support Vector Machine Combination Model. Journal of Artificial Intelligence Practice (2021) Vol. 4: 28-36. DOI: http://dx.doi.org/10.23977/jaip.2020.040206

REFERENCES

[1] Chen Jun, Tao Wei, LV Yingfei, he Jianping. Short term load forecasting based on Kalman filter [J]. Electrical switch, 2014 (2): 43-45
[2] Chen Heng, Wang Ke, Chen Lihua. Research on holiday short term load forecasting based on filtering algorithm [J]. Electrical technology, 2014 (9): 12-15
[3] Li minggan, sun Jianli, Liu Pei. Short term load forecasting of power system based on Kalman filter [J]. Relay, 2004 (4): 9-12
[4] I.Arasaratnam, S.Haykin. Cubature Kalman Filters[J]. IEEE Trans. Automat.Control, 2009, 54(6):1254-1269.
[5] Y.X.Wu,D.W.Hu,M.P.Wu,X.P.Hu. Quasi-Gaussian Particle Filtering[J].Comput. Sci., 2006, (3991):689-696.
[6] Li Yanhong, Lei Jinhui. Application and research of power load time series forecasting [J]. Science, technology and engineering, 2011 (04): 860-864
[7] Ji Xiaojun, Yang Juan. Application of neural network based on wavelet transform in short-term load forecasting of Shaanxi Power Grid [J]. Shaanxi electric power, 2013 (10): 66-70
[8] Zhao Zhiyong, Huang Wei, Yang Wei. A medium and long term power load forecasting method based on wavelet analysis and Elman dynamic neural network [J]. Shanxi electric power, 2013 (01): 1-5
[9] Li Xiao, Wang Xin, Zheng Yihui, Li Lixue, Sheng Xikui, Wu Hao. Short term wind power load forecasting based on improved least squares support vector machine and prediction error correction [J]. Power system protection and control. 2015 (11): 66-70
[10] HUANG F J,LECUN Y.Large-scale learning with SVM and convolutional for generic object categorization [C]. IEEE Computer Society Conference on Computer Vision and Pattern
[11] Recognition,2006: 284-291.
[12] Dong Hao, Cheng Peng, Li Lingling. Application of deep learning algorithm in short-term load forecasting of power system [J]. Electrical times, 2017 (02): 82-84

Downloads: 1212
Visits: 65253

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.