Time-varying reliability method applied to mechanical structure

Abstract

The invention discloses a time-varying reliability method applied to a mechanical structure. In consideration of the time variation of a mechanical structure in the time history, due to long-time use,time variation or dynamic characteristics often occur in parameter uncertainty and structural reliability due to the time variation generated by the material performance, the environment condition, the load effect and the like of the mechanical structure and the influence of various other factors; uncertainty of each parameter of an uncertainty random variable expression structure is introduced,and a Wiener process is used for describing a structural strength degradation process, so that the problem of calculation of the reliability of a time-varying structure is solved. On the basis of improving a primary second moment method, conversion from related non-normal variables to independent standard normal variables is completed through proposed linear Nataf changes, time-varying reliabilitymodeling is carried out again, and the reliability calculation problem that random parameters obey arbitrary distribution is solved. The method can effectively solve the problem of reliability and sensitivity during structural system strength degradation.

Description

technical field [0001] The invention belongs to the field of structural reliability analysis and relates to a time-varying reliability calculation method based on linearized Nataf transformation. Background technique [0002] Traditional reliability analysis methods do not consider dynamic uncertainties such as strength degradation and time-varying loads, so the reliability obtained is a constant value. For many engineering problems, due to long-term use, the time-varying nature of its own material properties, environmental conditions, load effects, etc., and other factors, parameter uncertainty and structural reliability often Time-varying or dynamic characteristics appear instead of a single value under traditional models. For the traditional static reliability model, reliability analysis only needs to be carried out at one moment, and the obtained reliability is a fixed value. The time-varying reliability model fully considers various time-varying uncertainties in engin...

AI Technical Summary

Problems solved by technology

At the same time, in actual engineering, random variables are generally related to non-normal distribution, and the problem will become more complicated

There are obvious defects in the commonly used methods at present, such as using the Rackwitz-Fiessler transformation for equivalent normalization, and then converting the relevant standard normal random variables into independent standard normal random variables through linear transformation (orthogonal transformation or triangular transformation) Variables, in the process of variable transformation, the correlation between the default variables does not change, or changes little, in fact, this assumption is only true in some special cases, for most distribution type transformations The difference in correlation between before and after is not negligible

Images