Evaluation For Machining Center Components Importance Based On Improved LeaderRank

Machining center is a complex equipment integration system.Artificial intelligence and system automation have increasingly higher requirements for complex equipment.The structure of the system becomes more and more complex,and the components are more and more diverse.The complex coupling relationship between components causes frequent fault propagation,which greatly reduces the reliability level of the whole machine.The manpower and material resources spent on fault maintenance cannot be ignored.Therefore,evaluating the importance of system components at different times is of great significance for reducing the occurrence of cascading failures and improving the reliability of the whole machine in operation.Firstly,the components were divided based on the structure and working principle of a certain machining center,and the fault propagation characteristics were analyzed combined with the fault information collected,and the SDG model was established.Matrix transformation was applied to realize the model hierarchy,and the fault stratified SDG model of the machining center was constructed,which laid a foundation for the subsequent component importance assessment of the machining center.Secondly,the fault distribution model is identified according to the AndersonDarling coefficient.D test method was used to test the goodness of fit,and the component reliability model was established.Coupla function is used to establish the joint reliability model of fault-related components,the GSS algorithm is used to estimate the correlation coefficient value and the SPI algorithm is selected for verification.Based on this,combined with the conditional probability theory and SDG model,the system component failure transmission probability model is constructed.Finally,assuming that the fault propagation obeys the Markov process,the LeaderRank algorithm is introduced and improved.The adjacency matrix was established according to the component failover probability and SDG model,and the component importance was evaluated by matrix iteration operation based on the state transition probability matrix.The advantages of the improved LeaderRank algorithm are verified by comparing with the evaluation results of node intermediate algorithm and Page Rank algorithm.Furthermore,the influence of time-varying fault transfer probability on the order of importance of components is analyzed.This paper has carried out in-depth research on the importance evaluation of machining center components.The proposed method solves the problem of identifying weak links in reliability under fault propagation and can also be used for fault source location.At the same time,the method presented in this paper is not only applicable to machining center products,but also can provide reference for the identification of key information sources in similar complex system information dissemination.