scholarly journals Proposing a global sensitivity analysis method for linear models in the presence of correlation among input variables

2016 ◽  
Vol 23 (2) ◽  
pp. 399-406 ◽  
Author(s):  
Y. Daneshbod ◽  
M. J. Abedini
Keyword(s):  
Sensitivity Analysis ◽  
Linear Models ◽  
Global Sensitivity Analysis ◽  
Analysis Method ◽  
Global Sensitivity ◽  
Sensitivity Analysis Method ◽  
Input Variables

Investigation of the uncertainty of the in-plane mechanical properties of composite laminates

2011 ◽  
Vol 226 (7) ◽  
pp. 1739-1750 ◽  
Author(s):  
Tian Longfei ◽  
Lu Zhenzhou ◽  
Hao Wenrui
Keyword(s):  
Mechanical Properties ◽  
Sensitivity Analysis ◽  
Composite Laminates ◽  
Global Sensitivity Analysis ◽  
Nonlinear Models ◽  
Analysis Method ◽  
Global Sensitivity ◽  
Response Variance ◽  
Sensitivity Analysis Method ◽  
Input Variables

The uncertainty of the in-plane mechanical properties of the laminate used in an aircraft wing structure is investigated. Global sensitivity analysis is used to identify the source of the uncertainties of the response performance. Due to the limitations of the existing global sensitivity analysis method for nonlinear models with correlated input variables, a new one using nonlinear regression is proposed. Furthermore, a contribution matrix is defined for engineering convenience. Two nonlinear numerical examples are employed in this article to demonstrate the ability of the proposed global sensitivity analysis method. After applying the proposed global sensitivity analysis method to the laminate model, the contribution matrices are obtained; from these matrices, researchers can identify the dominant variance contributions that contribute the most to the response variance. Factor analysis is then employed to analyze the global sensitivity analysis results and determine the most efficient methods to decrease the variances of the in-plane elastic constants. Monte Carlo simulation is used to demonstrate the efficiency of the methods in decreasing the variances.

A Global Sensitivity Analysis Method for Multi-Input Multi-Output System and its Application in Structural Design

2021 ◽  
pp. 2141005
Author(s):  
Qiming Liu ◽  
Nichen Tong ◽  
Xu Han
Keyword(s):  
Sensitivity Analysis ◽  
Global Sensitivity Analysis ◽  
Sensitivity Index ◽  
Response Functions ◽  
Analysis Method ◽  
Order Function ◽  
Global Sensitivity ◽  
Sensitivity Analysis Method ◽  
Output System ◽  
Input Variables

Commonly, variance-based global sensitivity analysis methods are popular and applicable to quantify the impact of a set of input variables on output response. However, for many engineering practical problems, the output response is not single but multiple, which makes some traditional sensitivity analysis methods difficult or unsuitable. Therefore, a novel global sensitivity analysis method is presented to evaluate the importance of multi-input variables to multi-output responses. First, assume that a multi-input multi-output system (MIMOS) includes [Formula: see text] variables and [Formula: see text] responses. A set of summatory functions [Formula: see text] and [Formula: see text] are constructed by the addition and subtraction of any two response functions. Naturally, each response function is represented using a set of summatory function. Subsequently, the summatory functions [Formula: see text] and [Formula: see text] are further decomposed based on the high dimensional model representation (HDMR), respectively. Due to the orthogonality of all the decomposed function sub-terms, the variance and covariance of each response function can be represented using the partial variances of all the decomposed function sub-terms on the corresponding summatory functions, respectively. The total fluctuation of MIMOS is calculated by the sum of the variances and covariances on all the response functions. Further, the fluctuation is represented as the sum of the total partial variances for all the [Formula: see text]-order function sub-terms, and the total partial variance is the sum of [Formula: see text] partial variances for the corresponding [Formula: see text]-order function sub-terms. Then, the function sensitivity index (FSI) [Formula: see text] for s-order function sub-terms is defined by the ratio of the total partial variance and total fluctuation, which includes first-order, second-order, and high-order FSI. The variable sensitivity index [Formula: see text] of variable [Formula: see text] is calculated by the sum of all the FSIs including the contribution of variable [Formula: see text]. Finally, numerical example and engineering application are employed to demonstrate the accuracy and practicality of the presented global sensitivity analysis method for MIMOS.

A global sensitivity analysis method based on ANOVA with convex set model and its state-dependent parameter solution

2018 ◽  
Vol 233 (11) ◽  
pp. 4039-4051
Author(s):  
Qing Guo ◽  
Yongshou Liu ◽  
Xiangyu Chen
Keyword(s):  
Sensitivity Analysis ◽  
Global Sensitivity Analysis ◽  
Convex Set ◽  
The State ◽  
Analysis Method ◽  
Global Sensitivity ◽  
State Dependent ◽  
Sensitivity Analysis Method ◽  
Dependent Parameter ◽  
Input Variables

Convex set model is most widely applied around nonprobabilistic uncertainty description. This paper combines the convex model with global sensitivity analysis theory of variance, and then proposes an index based on convex set model and variance-based global sensitivity analysis method to illustrate the effect of the nonprobability variables on the dangerous degree. The proposed index consists of two parts, including the main and total indices. The main index can quantitatively reflect the effect of uncertainties of input variables on the variance of output response, and the total index reflects the influence of interaction with other variables in addition to the individual influence of input variables. Furthermore, an efficient state-dependent parameter solution for solving the variance-based global sensitivity analysis of nonprobabilistic convex uncertainty is given in this paper. The state-dependent parameter solution not only greatly improves the efficiency but also guarantees the computational accuracy, and the times of performance functions evaluation decrease from [Formula: see text] in single-loop Monte Carlo solution to 2048 in the state-dependent parameter method. Finally, three numerical examples and a finite element example are used to verify the feasibility and rationality of the proposed method.

An analytical structural global sensitivity analysis method based on direct integral

2018 ◽  
Vol 27 (11) ◽  
pp. 1559-1576 ◽  
Author(s):  
Jie Liu ◽  
Longwei Tu ◽  
Guangzhao Liu ◽  
Chao Jiang ◽  
Zheng Zhang
Keyword(s):  
Sensitivity Analysis ◽  
Global Sensitivity Analysis ◽  
Analysis Method ◽  
Direct Integral ◽  
Global Sensitivity ◽  
Sensitivity Analysis Method
Sign in / Sign up

Export Citation Format

Share Document