Table of Content

Open Access iconOpen Access



A Lightning Disaster Risk Assessment Model Based on SVM

Jianqiao Sheng1, Mengzhu Xu2, Jin Han3,*, Xingyan Deng2

1 Information and Communication Branch of State Grid Anhui Electric Power Co., Ltd., Hefei, 230041, China
2 Information and Communication Branch of State Grid Shanxi Electric Power Co., Ltd., Taiyuan, 030000, China
3 School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, 210044, China

* Corresponding Author: Jin Han. Email: email

Journal on Big Data 2021, 3(4), 183-190.


Lightning disaster risk assessment, as an intuitive method to reflect the risk of regional lightning disasters, has aroused the research interest of many researchers. Nowadays, there are many schemes for lightning disaster risk assessment, but there are also some shortcomings, such as the resolution of the assessment is not clear enough, the accuracy rate cannot be verified, and the weight distribution has a strong subjective trend. This paper is guided by lightning disaster data and combines lightning data, population data and GDP data. Through support vector machine (SVM), it explores a way to combine artificial intelligence algorithms with lightning disaster risk assessment. By fitting the lightning disaster data, the weight distribution between the various impact factors is obtained. In the experiment, the probability of lightning disaster is used to compare with the actual occurrence of lightning disaster. It can be found that the disaster risk assessment model proposed in this paper is more reasonable for the lightning risk. It has been verified that the accuracy rate of the assessment model in this paper has reached 80.2%, which reflects the superiority of the model.


Cite This Article

J. Sheng, M. Xu, J. Han and X. Deng, "A lightning disaster risk assessment model based on svm," Journal on Big Data, vol. 3, no.4, pp. 183–190, 2021.

cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1049


  • 775


  • 0


Share Link