Reliability Assessment For Organic Coating-substrate With Failure Mechanism

As a typical long-term storage product,hydrogen storage equipment is of great significance to the development of hydrogen industry.Coating-substrate structure is a key structure of hydrogen storage equipment.Accurately evaluating its storage reliability is of vital importance to guarantee reliability and safety of hydrogen storage equipment.On the one hand,degradation phenomena such as cracking and bubbling of organic coating lead to water penetration,and the substrate will corrode with water over time.Because the processes of coating degradation and substrate corrosion are dynamic and correlated,it is difficult to accurately characterize the degradation of structure by the traditional methods.On the other hand,due to the microscopic variations,there exists spatial variability for the organic coating,resulting in the non-uniform corrosion of the substrate,and it is,therefore,difficult to characterize the spatial variability by the traditional uniform corrosion.In this dissertation,based on the failure mechanism of substrate and organic coating,the multi-physics model of substrate and organic coating-substrate is carried out,and the reliability evaluation method of the organic coating-substrate is studied by the multiphysical simulation model.Additionally,considering the spatial variability of coating property parameters,a reliability evaluation method with organic coating-substrate is proposed.The primary research contributions are as follows:(1)Development of a multi-physics simulation method for substrate corrosion based on failure mechanism.In this dissertation,based on the corrosion mechanism of lithium hydride,a multi-physics simulation model is carried out by the COMSOL Multiphysics.Besides,the influence of different temperatures and relative humidity on the substrate corrosion is studied.The results show that substrate corrosion occurs gradually from the outside to the inside.Furthermore,the greater the temperature and the relative humidity,the faster the substrate corrodes.(2)Development of a multi-physics simulation-based method for the reliability evaluation of organic coating-substrate based on failure mechanism.A multi-physics simulation model is established with the mechanism of substrate corrosion and the water penetration of coating.By taking account of the multi-physics simulation model,the sensitivity analysis of the input parameters is carried out.We found that the porosity of coating is the crucial factor to the substrate corrosion.By using the regression model with the power function,the relation between the inputs and outputs of multi-physics simulation model is characterized.With the assumption that the degradation of organic coating follows a Gamma process,the reliability of organic coating-substrate is obtained by a regression model.(3)Development of a reliability evaluation method for organic coating-substrate considering spatial variability of organic coating parameters.The spatial variability of coating parameters caused by the microscopic variations on a surface of organic coating is quantified by a random field.By comparing different random field discrete methods,we found that the Expansion Optimal Linear Estimation(EOLE)method outperforms the K-L series expansion method.Considering spatial variability of organic coating parameters,the multi-physics simulation model of organic coating-substrate is built by a random field.A Convolutional Neural Network(CNN)is utilized as the metamodel to characterize the relation between the inputs and outputs of the multi-physics simulation model.Finally,the reliability analysis of organic coating-substrate considering the spatial variability of coating parameters is carried out.The results show that the reliability of the structure with the spatial variability of coating parameters is much lower than that without considering the spatial variability.