Analysis Of The Dynamic Reliability Of Structure And System Based On The Probability Density Evolution Method

In the reliability analysis of mechanical structure,the input load,failure mode and structural feature functions of mechanical equipment are random in time domain,space and intensity,which correspondingly results in unpredictable nonlinear state including reliability prediction.To solve the nonlinear equations are challengeable and difficult.Therefore,it is very important to study on the nonlinearly stochastic dynamical system with the coupling of randomness and nonlinear factors.A new dynamic reliability modeling is proposed based on the study of dynamic reliability of parts under repeated loads and the theory of probability density evolution.A new calculation technique for the dynamic reliability of mechanical system is then discussed,and the security and risk assessment to the system when considering various factors and the redundant algorithm optimization are also investigated.The main content of this dissertation is as follows:1.The reliability research is reviewed and the work direction and target problem to be solved are demonstrated.A new analysis model for the dynamic reliability is presented by combining the existing dynamic reliability model with the probability density evolution(PDEM),and the model is verified with a numerical solution.2.The traditional reliability models of mechanical systems mostly follow the computational method of the reliability of electronic components in electronic circuits,that is,the reliability of the systems will be simply computed based on the connection logic of the components.A dynamic reliability model for the failure-dependent mechanical system is proposed when adopting copula function to deal with the dynamic correlation existing in the parts' failure modes in order to establish a more accurate and reasonable analysis method for the system reliability.3.Because complex equipment systems are typically integrated systems of electronics,machinery and control,which are composed of structural components,electronic components,control components and electrical equipment,etc.Each unit or component in the whole system works differently and also influence differently the reliability of the whole system.In this work,analytic hierarchy process(AHP)will be used for hierarchical and structured processing of complex systems,and the multilayer reliability evaluation system is then constructed so that the designers can identify weaknesses of the system more effectively,treat each unit and its impacts rationally,and carry out reliability optimization design work.4.Redundant design has the advantages such as the low cost and remarkable effect and the reliability of the whole system can be guaranteed by adding up to spare parts with this design.Theoretically,the failure rate will infinitely approach to zero with the increasing spare parts.On the other hand,it is also very important to strictly control the cost,installation volume and weight of the systems.Such problems above can be simplified as a mathematics problem of multi-objective algorithm optimization with boundary conditions.with the increasing optimization objectives and complicating dynamic environment,the accuracy of optimization results will be tested out.Here,based on the redundancy design of mechanical system,a new algorithm aiming at finding more accurate results is proposed,and the comparison verification is then carried out.