Abstract
Deep mining of power cable fault information can improve the analysis of cable fault influencing factors. Therefore, based on the 10 kV power cable fault data of a power supply company, the statistical model—Cox proportional hazard model was used to quantitatively analyze the influencing factors of cable faults to guide cable procurement, construction, operation and maintenance. In order to ensure the accuracy of data analysis, the cable data preprocessing principle is proposed, and the appropriate sample size is discussed. The Cox proportional hazards model was used to analyze the influencing factors of cable faults; the Logistic regression model was used to determine the types of influencing factors of cable faults, and the cable failure rate corresponding to each influencing factor of cable faults was calculated statistically, and the components of each influencing factor were determined. The relative risk degree finally proved the correctness of the Cox proportional hazards model analysis results. The results show that the cable failure rates of main body manufacturer M1, accessory manufacturer N1, and construction unit I3 are 0.33, 0.29, and 0.218 respectively. Enterprises should pay attention to these three units when purchasing, constructing, and maintaining cables.
Original language | English |
---|---|
Pages (from-to) | 2442-2450 |
Number of pages | 9 |
Journal | Gaodianya Jishu/High Voltage Engineering |
Issue number | 8 |
DOIs | |
Publication status | Published - 31 Aug 2016 |
Keywords
- power cable
- Fault influencing factorsmodel
- sample
- Cox proportional hazard
- Logistic regression model
- cable failure rate
ASJC Scopus subject areas
- Energy Engineering and Power Technology
- Electrical and Electronic Engineering