uq & sensitivity analysis - for correlated variables · 12/18 inpractice,... i...

20
UQ & Sensitivity analysis for correlated variables Anne Eggels CASA day April 19, 2017

Upload: others

Post on 23-May-2020

8 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

UQ & Sensitivity analysisfor correlated variables

Anne Eggels

CASA dayApril 19, 2017

Page 2: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

1/18

Uncorrelated data

For uncorrelated data,

Independent variables ⇒ stochastic collocation

Number of nodes k grows exponentiallywith number of variables p.

Page 3: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

2/18

Wind and wave data

Our data is correlated!

Dependent variables⇒ clustering-based collocation (CBC)

Page 4: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

3/18

ResultI The method was tested by integrating test

functions over different data sets.

I Higher correlation leads to smaller error.

Example with 50 nodes and 105

data points

Page 5: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

4/18

What now?

Collocation points and weights are known.

Output can be computed.

What to do with the output?

Page 6: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

5/18

Goals of sensitivity analysis (1/2)

Increased understanding of the relationships betweeninput and output variables in a system or model

I Which input variables influence the output and howmuch?

I How to detect/quantify correlations between inputvariables?

I Causality

Page 7: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

6/18

Goals of sensitivity analysis (2/2)

Improved uncertainty analysis to better estimate theoutput probability distribution function

I More weight to more important variables

I What to do with groups of correlated variables?

Page 8: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

7/18

Example - x,y ∼ U[0, 1], f(x,y) = f(x) = x2

Page 9: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

8/18

Adaptive CBC

I Apply CBC with a fraction of the computationalbudget

I Apply SA and determine important variables

I Re-apply CBC with the rest of the budget on onlythe important variables

I Iterate?

Remark: if iterations are used, then variables can notbe added again.

Page 10: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

9/18

Example for SA - minimum spanning trees

L = 0.0995 L = 3.3695

Page 11: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

10/18

Reference distribution

Page 12: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

11/18

Method for SA - minimum spanning trees

I Length of the minimum spanning tree is an inversemeasure for correlation

I Include the weights of the clusters

I Compare length to reference length

I Use only “important” variables in the collocationmethod

Page 13: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

12/18

In practice, ...

I we would like to compute the MST on the completedataset

I output data is far from complete

Solution:

I Interpolate the output data

I Use the previous method based on CBC forinput/output-relations

I Use a multilevel method (with corrections) forinput/input-relations

Page 14: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

13/18

Mutual information

I Measure of mutual dependence between tworandom variables

I “Amount of information” obtained about onerandom variable by knowing the other

I I(X,Y) = HS(X) +HS(Y)−HS(X,Y)

I HS(·) Shannon entropy (measure of information /unpredictability)

I Easy for distributions, hard for datasets

Page 15: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

14/18

Minimum spanning trees and mutualdependence

I α-entropy:

Hα(f) =1

1− αlog

(∫Ωfα(ω)dω

)for α ∈ (0, 1)

I Length of MST converges (within a constant) to the1/2-entropy!

log(

LNβ√N

)→ H1/2

I Hα(X) = Hα(Y) for all X,Y =⇒ Hα(X,Y) measure fordependence

Page 16: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

15/18

How to handle groups of important variables?

Clique idea (future work)

Page 17: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

16/18

Testcase - wind op zee

Dataset: OWEZ (Offshore Windpark Egmond aan Zee)

Model: wake model for Horns Rev

Real input does not match restrictionsof model on input data!

Page 18: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

17/18

Current and future work

I How to handle groups of important variables?(clique idea)

I How to determine which variables are importantenough?

I How to allocate simulation budget to first andsecond collocation?

I Real-life tests

Page 19: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

18/18

Conclusions

I We developed a new method for sensitivity analysiswhich works for correlated variables.

I The method has a theoretical basis, but detailshave to be worked out.

I When combined with uncertainty analysis, this canlead to more insight in complex problems.

Page 20: UQ & Sensitivity analysis - for correlated variables · 12/18 Inpractice,... I wewouldliketocomputetheMSTonthecomplete dataset I outputdataisfarfromcomplete Solution: I Interpolatetheoutputdata

1/1

Questions?

This research is supported by the Dutch Technology Foundation STW, whichis part of the Netherlands Organisation for Scientific Research (NWO), and

which is partly funded by the Ministry of Economic Affairs.