摘要
建立了Weibull分布和Crow-AMSAA模型用来分析电缆故障数据以预测电缆故障率和指导维修更换;数学推导表明,两参数Weibull分布和Crow-AMSAA模型的表达式具有一致性,Crow-AMSAA模型是一种特殊形式的两参数Weibull分布.对比Weibull分布和Crow-AMSAA模型对某供电公司电缆8年故障数据的分析结果表明,Weibull分布可定性分析失效时间,可以得到故障的模式;Crow-AMSAA模型可定量分析累积故障次数和累积故障时间,可预测未来故障次数;当故障数据样本较小时,Weibull分布的分析结果相对准确.
This paper has built Weibull distribution and Crow-AMSAA model to analyse the power cable failure data so as to predict the failure rate.It is found through the mathematical deduction that the Crow-AMSAA model is a special form of the two parameters Weibull distribution.By using Weibull distribution and Crow-AMSAA model,the analysis results of 8 years failure data of a power supply company are compared.It is found that the Weibull distribution is qualitative in nature.It uses age to failure data as input,from which failure model can be obtained.In comparison,the Crow-AMSAA model is quantitative.It takes on cumulative time vs cumulative faults data from which the future number of failure can be acquired.When the size of failure data sample is small,the result of Weibull distribution is more reliable.
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2012年第5期667-672,共6页
Engineering Journal of Wuhan University
基金
Engineering and Physical Sciences Research Council(UK)(EP/G028397/1)
