Evaluation Method of Switchgear State Based on Adaptive DBSCAN Algorithm
2020 5th Asia Conference on Power and Electrical Engineering (ACPEE)(2020)
摘要
In order to better identify the outlier in the detection data and reasonably evaluate the state of the switchgear, we propose a method for evaluating the state of switchgear based on the Density-Based Spatial Clustering of Applications with Noise (DBSCAN). First, the feature quantities of the switchgear is composed of the insulation state index , the Transient Earth Voltage (TEV), the Ultrasonic Testing (UT), the ambient temperature, the ambient humidity and the operation life. Standardization method is carried out for all feature quantities to form multi-dimensional feature dataset .Then the k-average nearest neighbor method and mathematical expectation method are used to generate the R radius and the minimum number of MinPts parameters. The density threshold (Den) is introduced to automatically find the stable range of cluster number. Finally, In this paper, the field test data is used as an example to verify the feasibility of the method and provide a theoretical basis for the evaluation of switchgear.
更多查看译文
关键词
switchgear,state evaluation,adaptive optimization,DBSCAN algorithm,outlier
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络