A Predicting Method of Power Grid Transmission Line Icing Based on Decision Trees Modified Model

Abstract:

Aiming at collapse, disconnection, and ice flash-over problems caused by power grid transmission line icing, this paper proposes a predicting method of transmission line icing. According to historical data and theories of decision tree, calculating out each factors information entropy, with an overall consideration of each influencing factors attribute and attribute value and connections between adjacent layers, we pick out the best testing factor as the root note to complete the model. This model has a clear train of thought, at the same time it has high predicting accuracy and efficiency, which provides a strong guarantee for safely running of power grid transmission line.

• Periodical: Applied Mechanics and Materials (Volumes 347-350)
• Pages: 1903-1906
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.347-350.1903
• Online since: August 2013
• Authors: Jian Wang, Wei Qing Ma, Xiao Long Zhao, Fei Fei Liang
• Keywords: Accuracy, Decision Tree, Icing, Information Entropy, Power Grid

References

[1] Du Yunying, Xing Jun, et al. A reason and strategy research of southern icing disaster, Jiangsu electrical engineering, 2008, 27(6): 14-17.

[2] Yan Tongxi, et al. Transmission line icing meteorological parameter analysis, Mechanical management and development, 2006, (5): 51-52.

[3] Pi Zhixian, Power load forecasting research based on decision tree, North China electric power university master degree thesis, (2011).

[4] Liao Xianglin, Transmission line icing property classification and density, Electric power construction, 1994, (9).

[5] Tan Guanri, several microclimate discussion on transmission line icing [J], Acta Meteorologica Sinica, (1982).