Importance Sampling-based Post-Processing Method for Global Sensitivity Analysis

Sparkman, Daniel; Millwater, Harry; Garza, Jose; Smarslok, Benjamin P.

doi:10.2514/6.2016-1440

Cited by 6 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, if

f_{Y, X_{i}} (y, x_{i})

accurately describes the joint PDF of the input‐output sample, we can simply compute various sensitivity indices based on Equation . This has been motivated a number of studies to investigate several strategies for characterizing the joint PDF of the input‐output data for the purpose of variance‐based GSA (see, e.g., Cheng et al., 2015; Jia & Taflanidis, 2016; Sparkman et al., 2016; Wei et al., 2015).…”

Section: Methodsmentioning

confidence: 99%

“…The method is mainly designed to alleviate the high‐computational demand associated with GSA of process‐based hydrologic models. At present, unlike the sampling‐based GSA methods, most of the given‐data methods only focus on estimating the variance‐based first‐order sensitivity index (i.e., main effect) and cannot estimate the total sensitivity and interaction effects (see, e.g., Sparkman et al., 2016; Strong & Oakley, 2013; Strong et al., 2012). Hence, the secondary objective of this paper is to develop an efficient data‐driven method to simultaneously compute first‐order and total effect indices.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Sheikholeslami

Gharari

Papalexiou

et al. 2021

Water Resources Research

View full text Add to dashboard Cite

Introduction On the High Computational Cost Incurred by the Sampling-Based Global Sensitivity AnalysisWith the rapid increase in computational capability, an increasing number of distributed and semi-distributed process-based hydrologic models have been introduced to simulate the quantity and quality of water on a range of spatiotemporal scales. The ever-growing complexity of these models is driven by the

show abstract

“…Therefore, if

f_{Y, X_{i}} (y, x_{i})

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Sheikholeslami

Gharari

Papalexiou

et al. 2021

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…where Var X i (Y|X ~i) represents the variance of Y when all variables other than X i are fixed at specific values, and E X ~i calculates the expected value of this variance considering the randomness in X ~i. Some techniques to compute these sensitivity indices are the Sobol's scheme (Sobol′, 2001), Fourier amplitude sensitivity test (FAST) (Saltelli et al, 1999), improved FAST (Tarantola et al, 2006), importance sampling and kernel regression (Sparkman et al, 2016), and the stratified sample-based approach for sensitivity analysis (Li and Mahadevan, 2016). Here, we use the stratified sample-based approach for its computational efficiency.…”

Section: Variance-based Sensitivity Analysismentioning

confidence: 99%

Online monitoring and control of a cyber-physical manufacturing process under uncertainty

et al. 2020

View full text Add to dashboard Cite

Recent technological advancements in computing, sensing and communication have led to the development of cyber-physical manufacturing processes, where a computing subsystem monitors the manufacturing process performance in real-time by analyzing sensor data and implements the necessary control to improve the product quality. This paper develops a predictive control framework where control actions are implemented after predicting the state of the manufacturing process or product quality at a future time using process models. In a cyber-physical manufacturing process, the product quality predictions may be affected by uncertainty sources from the computing subsystem (resource and communication uncertainty), manufacturing process (input uncertainty, process variability and modeling errors), and sensors (measurement uncertainty). In addition, due to the continuous interactions between the computing subsystem and the manufacturing process, these uncertainty sources may aggregate and compound over time. In some cases, some process parameters needed for model predictions may not be precisely known and may need to be derived from real time sensor data. This paper develops a dynamic Bayesian network approach, which enables the aggregation of multiple uncertainty sources, parameter estimation and robust prediction for online control. As the number of process parameters increase, their estimation using sensor data in real-time can be computationally expensive. To facilitate realtime analysis, variance-based global sensitivity analysis is used for dimension reduction. The proposed methodology of online monitoring and control under uncertainty, and dimension reduction, are illustrated for a cyber-physical turning process.

show abstract

Efficient global sensitivity analysis with correlated variables

DeCarlo

Mahadevan

Smarslok³

2018

Struct Multidisc Optim

View full text Add to dashboard Cite

Importance Sampling-based Post-Processing Method for Global Sensitivity Analysis

Cited by 6 publications

References 8 publications

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Online monitoring and control of a cyber-physical manufacturing process under uncertainty

Efficient global sensitivity analysis with correlated variables

Contact Info

Product

Resources

About