Open AccessResearch on Early Warning of Power Grid Construction Safety Based on PSO-SVM Model
Author(s) -
Yuwei Sun,
Jianguo Zhu,
Xinghua Yu,
Chao Ye,
Guoxiang Fan
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1449/1/012037
Subject(s) - particle swarm optimization , warning system , support vector machine , computer science , power grid , electric power , electric power system , grid , early warning system , reliability engineering , power (physics) , operations research , engineering , artificial intelligence , machine learning , telecommunications , physics , geometry , mathematics , quantum mechanics
Electric power system is an important symbol of China’s economic development, and electric power construction is an important way of electric power system. However, the potential safety hazards and safety accidents generated during its development have a great impact on economic and social development. In order to reduce the occurrence rate of such accidents, this paper proposes a power grid construction safety early warning model based on PSO-SVM. The quantitative model adopted by the traditional SVM safety early warning method has limitations in parameter optimization. The particle swarm optimization (PSO) algorithm is superior to the traditional method in parameter optimization because it cannot obtain better early warning effect. It uses the information sharing between the whole group and the mutual cooperation between individuals to search, thus searching for better parameter combination and achieving better early warning effect. Finally, the traditional SVM model test for power grid construction is compared, further proving the improvement of the model after improvement.