Title :
Condition-Based Maintenance of Electrical Equipments Based on 1D3 Algorithm and Monte Carlo Stimulation
Author :
Ran, Li ; Li, Chen ; Xueping, Gu
Author_Institution :
Key Lab. of Power Syst. Protection & Dynamic Security Monitoring & Control, North China Electr. Power Univ., Baoding
Abstract :
Condition-based maintenance is an advanced maintenance strategy which is based on the monitoring of equipments status and according to the results of monitoring and analysis people can arrange maintenance scientifically. When monitoring and analyzing the status of equipments, there are vast historical and on-line data to analyze. If we can find the main factors leading to equipment failure through analyzing the data and pay more attention to these factors, we can make maintenance decision more exactly and quickly. In this paper we propose to apply ID3 algorithm to classify the data according to the importance of their influence to equipment failure and then find the factors which influence equipments most. After finding the most important factors, apply Monte Carlo simulation to forecast the probability of equipment failure which occurs in their operational life span, and according to this we can pay more attention to monitor the factors which result in failure easily
Keywords :
Monte Carlo methods; decision trees; maintenance engineering; power apparatus; probability; ID3 algorithm; Monte Carlo simulation; condition-based maintenance; data mining; equipment failure; equipments status monitoring; operational life span; probability; Algorithm design and analysis; Condition monitoring; Data analysis; Data mining; Equipment failure; Failure analysis; Maintenance; Monte Carlo methods; Power system reliability; Time series analysis; Condition-Based Maintenance; Data Mining; ID3 Algorithm; Monte Carlo Stimulation;
Conference_Titel :
Transmission and Distribution Conference and Exhibition: Asia and Pacific, 2005 IEEE/PES
Conference_Location :
Dalian
Print_ISBN :
0-7803-9114-4
DOI :
10.1109/TDC.2005.1547029