Reliability Modeling And Assessment For Long-storage Systems Based On Degradation Data

Long-storage equipment,which has the characteristics of “long-term storage one-time use”,is an extremely important system and of great significance to national security and national defense construction.Compared with general products and industrial technical equipment,long-storage equipment needs to be stored for a long time.During the storage time,equipment must stay with high reliability to ensure that the emergency tasks can be successfully completed.Nevertheless,in practical engineering,due to the impact of complex environmental stresses,the critical structure of long-storage equipment will undergo aging and degradation,resulting in continuous degradation of its performance.The mechanism that causes the performance degradation and failure of the critical structure is complicated,and it is difficult to directly establish an effective physical-based failure model to reveal the performance degradation process.Meanwhile,the lack of sufficient lifetime data makes it difficult to apply traditional reliability assessment methods.Hence,how to accurately characterize the degradation behavior of the critical structure in the storage stage and establish an effective reliability assessment model is a research challenge in the field of reliability engineering.To study the degradation behavior and reliability assessment of the critical structure of long-storage equipment in the storage stage,it is necessary to grasp the performance degradation trend,so as to establish an effective reliability assessment model.Based on the accelerated degradation and simulation data of long-storage equipment,this dissertation devotes to developing the modeling and evaluation methods for the critical structure of long-storage equipment.The research works are summarized as follows:(1)Evaluation the reliability of seal ring in long-storage equipment by the double-stress accelerated degradation data.Aiming at the difficulty of acquiring real data of long-storage equipment components,this dissertation uses accelerated degradation data,combined with single-factor and double-factor acceleration models,to establish random degradation model for components.The maximum likelihood estimation method is used to infer the unknown parameters in the model,and the reliability of the components under the consideration of single acceleration factor and double acceleration factors is compared.A common long-storage component seal ring is taken as an illustrative example.By comparing the reliability of the seal ring under different model,we found that it is more reasonable to consider the two-factor acceleration and adopt the stochastic degradation modeling method.(2)Development of a system reliability modeling method for the silicon-containing foam sandwich structure based on degradation data.In view of the silicon-containing foam sandwich structure,this dissertation establishes a finite element model of the sandwich structure with the aid of ANSYS.The orthogonal experimental design method is used to select the component-level degradation feature.For the sake of modeling the relationship between component’s degradation features and structural degradation features,this dissertation uses a MATLAB-ANSYS joint coding framework.By constructing the limit state function of the structure,regression methods is employed.Finally,on the basis of this equation,the degradation data of the preload between layers is simulated.(3)Implementation of the multi-level information aggregation method to carry out system reliability assessment of the silicon-containing foam sandwich structure.Under the Bayesian-based reliability information aggregation framework,this dissertation constructs the system reliability model with multi-level information aggregation.The cumulative hazard increments are constructed to serve as the basis for aggregating reliability information from lower-level elements toward higher-level elements.The Markov Chain Monte Carlo is applied to calculate the posterior distribution of cumulative hazard increments.The proposed method is used for the reliability assessment of the silicon-containing foam sandwich structure.By comparing the results obtained by the traditional statistical analysis method and the proposed method,the effectiveness and accuracy of the proposed method were examined.