Reliability analysis of deepwater explosion test vessel based on dynamic prediction

A deep-water explosion test vessel is an important test equipment which is filled with water to simulate different water depth environment by loading different hydrostatic pressure, and it can be used to study deep water explosion theory and engineering technology based on the similarity principle. In order to ensure the safety of vessels used in deep-water explosion test, a random-interval dynamic reliability model based on intelligent prediction is proposed in this paper. A GRNN prediction network of vessel response is established through dynamic test data, and the maximum strain interval variable of the vessel is obtained. Considering the random characteristics of the vessel structure, the reliability analysis of the in-service deep-water explosion test vessel is carried out. During the period, three methods are used to calculate the reliability index, and the analysis shows that for the vessel structure with high reliability and lack of sample data, the hybrid reliability model based on dynamic prediction is established by the calculation of interval reliability index. The method is simple and feasible. At the same time, the interval variables of the model change with the structural dynamic test data, and the uncertainty analysis of the structure is also dynamic. Therefore, the reliability of the container obtained is also changing with the service process, and has dynamic characteristics, which can better reflect the performance changes of the container during the service period and provide the basis for the use and maintenance decision of the container. © 2021, Editorial Staff of EXPLOSION AND SHOCK WAVES. All right reserved.