An Efficient Method for Multimodal Stochastic Uncertainty Analysis

Abstract

The present invention proposes an efficient stochastic uncertainty analysis method that can be used to process multi-peak distributed random variables by combining the univariate dimensionality reduction method (UDRM) and the maximum entropy method (MEM). This method extends MEM from the fourth-order moment constraint to the n-order moment constraint, and then determines the order of the response statistical moment (or the moment constraint of MEM) when the response distribution converges by performing UDRM+MRM cycle, and on this basis, uses The UDRM+MEM method calculates the probability distribution of the response and the probability of each point of the response. The present invention can simultaneously deal with multimodal distribution random variables and multimodal distribution responses, thereby better solving the problem that the Jacobian matrix G is close to singularity or ill-condition, and increasing the solvability of equations; at the same time, on the premise of ensuring the accuracy of results Under this condition, the demand for sample size can be greatly reduced, and a high calculation accuracy can be obtained with only a very small number of samples.

Description

technical field [0001] The invention belongs to the field of random uncertainty analysis methods and relates to an efficient multi-peak random uncertainty analysis method. Background technique [0002] In engineering, it is hoped that the product will be reliable, stable and safe, but various uncertain factors such as manufacturing errors, changes in material properties, changes in the use environment, and incomplete understanding will adversely affect product quality. It is of great significance to effectively measure and control these uncertainties to ensure the quality and reliability of products. Uncertainty is usually divided into two types: random uncertainty and cognitive uncertainty. Random uncertainty, also known as objective uncertainty, comes from the inherent randomness and volatility of physical systems or environments. Uncertainty that is eliminated with the improvement of cognitive level is also the research object of the present invention. [0003] There ar...

AI Technical Summary

Problems solved by technology

If the traditional method is used, it is difficult to obtain better calculation accuracy due to its own limitations, and it is also difficult to accurately estimate the bimodal characteristics of the response

[0004] In view of the above-mentioned problems in the traditional method of dealing with multimodal stochastic uncertainty and the lack of research on multimodal stochastic uncertainty, it is necessary to propose an efficient stochastic uncertainty analysis for dealing with multimodal random variables method

Images