Structural reliability analysis method based on agent model under condition of hybrid uncertainty

Abstract

The invention discloses a structural reliability analysis method based on an agent model under the condition of hybrid uncertainty. The structural reliability analysis method comprises the steps that random uncertainty is modeled by adopting random variables, and epistemic uncertainty is modeled by adopting interval variables; uniform samples generated by the random variables in an approximate value section and all the interval variables within upper bounds and lower bounds of respective intervals and a system response value are used as training sample points for establishing a Kriging agent model, and according to the obtained Kriging agent model, the maximum values and the minimum values of the reliability sensitivity, the failure probability and the system failure probability under any interval variable value are calculated by adopting a Monte Carlo simulation method. By the adoption of the structural reliability analysis method, the problems that a traditional agent model under the condition of the hybrid uncertainty has a certain precision locally, and the traditional reliability analysis calculated quantity is larger are solved, and the structural reliability analysis method better accords with engineering practice.

Description

technical field [0001] The invention belongs to the technical field of reliability analysis and evaluation, and relates to a reliability analysis method for mechanical products, in particular to a new method for structural reliability analysis based on a proxy model under mixed uncertainties. Background technique [0002] With the rapid development of science and technology, the structure of many products (such as: aircraft, aero-engines, deployable antennas, automobiles, CNC machine tools, etc.) has become more and more complex, and its analysis and design involve many disciplines and new methods and processes , Application of new materials. Due to the high cost of some products and the harsh working environment, if there is a failure during operation, it will cause huge economic losses and casualties. For example, the engine is the heart of the aircraft, known as the "flower of industry", it directly affects the performance, reliability and economy of the aircraft, and is...

AI Technical Summary

Problems solved by technology

It should be pointed out that the existing structural reliability analysis based on agent models only considers random uncertainties, but cannot effectively solve cognitive uncertainties.

In addition, the existing methods are greatly affected by the experimental points, and the established surrogate models can only have a certain accuracy locally, but cannot effectively approximate the entire uncertainty space.

Therefore, different test points often lead to different results, and the error is large, which seriously affects its effectiveness.

Images