Life-time Distribution Reduction Of Multi-component System And Preventive Maintenance Strategy Based On Parameter Uncertainty

With the increasing of huge loss due to failure maintenance for equipments,results in the higher requirement of reliability.Preventive maintenance method is one of most important method for improve reliability to the systems or components.Based on Monte Carlo simulation,the preventive maintenance optimization model for multi-component system is established,according to a case study of a hydraulic system(such as ZCL-1600 drilling mud pump pressure cylinder hydraulic system)of preventive maintenance optimization problem,we analyzed the applicable range of system when adopt preventive maintenance strategy.The lifetime distribution of a multi-component series system is Poly-Weibull distribution when Weibull parameters of each component are inconsistent.However,the increasing of component number in system results in the increasing of computational complexity when dealing with preventive maintenance optimization.To solve this problem,it is proposed a kind of double approximate fitting novel method to reduce the Poly-Weibull distribution into Single-Weibull distribution.This method contains two steps,first is to make an approximate distribution fitting based-on input performance,second is to make an approximate test based-on output performance.According to the results of numerical cases study for certain hydraulic system,it is proved that the double approximate fitting method is valid in certain range of system parameters.However,when approximate fitting the life-time distribution for multi-component,there may exist uncertainty of parameters or model.So we presented a method to analyze the influence of nonidentical model parameters on preventive maintenance strategy.Firstly,the uncertainty of parameters is quantitatively determined,then the model is established to analyze the influence of parameters uncertainty on the optimal maintenance interval and the optimal maintenance cost of the system.The simulation result showed that the uncertainty of model parameters has an important influence on the preventive maintenance decision.If the uncertainty is taken into account,it will make preventive maintenance of system more economical and improve the system reliability to some extent.