Adaptive Surrogate Modeling for Response Surface Approximations with Application to Bayesian Inference Serge Prudhomme a and C. Bryant b a Department of Mathematics, Ecole Polytechnique de Montr´ eal SRI Center for Uncertainty Quantification, KAUST b ICES, The University of Texas at Austin, USA Advances in UQ Methods, Algorithms, and Applications KAUST, Thuwal, Saudi Arabia January 6-9, 2015 S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 1 / 35
35
Embed
Adaptive Surrogate Modeling for Response Surface ... workshop... · I Ignore physical discretization error, pseudo-spectral projection, improved linear functional S. Prudhomme and
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Adaptive Surrogate Modeling for Response SurfaceApproximations with Application to Bayesian Inference
Serge Prudhommea and C. Bryantb
aDepartment of Mathematics, Ecole Polytechnique de MontrealSRI Center for Uncertainty Quantification, KAUSTbICES, The University of Texas at Austin, USA
Advances in UQ Methods, Algorithms, and ApplicationsKAUST, Thuwal, Saudi Arabia
January 6-9, 2015
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 1 / 35
Outline
Outline
• Introduction.
• Error estimation for PDEs with uncertain coefficients.
• Adaptive scheme.
• Numerical examples.
• Application to Bayesian Inference.
“. . . It is not possible to decide (a) between h or p refinement and(b) whether one should enrich the approximation space Vh or Sh
. . . better approaches, yet to be conceived, are consequently needed.”
Spectral Methods for Uncertainty Quantification,Le Maıtre & Knio 2010
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 2 / 35
Introduction
Introduction
A(λ;u) = f(λ) → Q(u(λ))︸ ︷︷ ︸M(λ)=Q(u)
λ1
λ 2
QN
(u)
Surrogate modelMN
M≈MN (λ) = Q(uN )
Ah(λ;uh) = fh(λ) → Q(uh(λ))︸ ︷︷ ︸Mh(λ)=Q(uh)
λ1
λ 2
Qh,N
(u)
Surrogate modelMh,N
Mh ≈Mh,N (λ) = Q(uh,N )
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 3 / 35
Introduction
References
Le Maıtre et al., 2007, 2010I Polynomial chaos, Stochastic Galerkin, Burger’s equation
Almeida and Oden, 2010I convection-diffusion, sparse grid collocation
Butler, Dawson, and Wildey, 2011I Stochastic Galerkin, PC representation of the discretization error
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 14 / 35
Numerical results
Example 2: Response surface with discontinuity
Convection-diffusion model in 2D:
−2∆u+
[sin(3π
2 ξ1)4bξ2 − ξ1c
]· ∇u = f(ξ) in D = (0, 1)2
u = 0 on ∂D
Loading f is chosen so that
u(x, y, ξ) = 10 sin(3π
2ξ1
)(4bξ2 − ξ1c
)· (x− x2)(y − y2)
where
bξ2 − ξ1c =
0 ξ1 ≤ ξ2
−1 ξ1 > ξ2
with ξ1, ξ2 ∼ U(0, 1).
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 15 / 35
Numerical results
Example 2: Response surface with discontinuity
Q(u(·, ξ)) = u
(1
3,1
3, ξ
)≈∫Dq(x, y) u(x, y, ξ) dxdy
q(x, y) =100
πexp
(− 100(x− 1/3)2 − 100(y − 1/3)2
)
00.2
0.40.6
0.81
0
0.5
1
−2
−1
0
1
2
ξ1
ξ2
True response for QoI overparameter space.
Adaptive scheme:
If ‖EΩi‖L2Ωi
> 0.75 maxj∥∥EΩj
∥∥L2
Ωj
split Ωi into 2n new elementsby bisection in each stochasticdirection
end
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 16 / 35
Numerical results
Example 2: Response surface with discontinuity
Adaptive hΩ refinement
0 0.2 0.4 0.6 0.8 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Total error
ξ1
ξ 2
−0.5
−0.4
−0.3
−0.2
−0.1
0
0.1
0.2
0.3
0.4
0.5
0 0.2 0.4 0.6 0.8 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Parameter space mesh
ξ1
ξ 2
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 17 / 35
Numerical results
Example 2: Convergence of true error
100 101 102 103 104
number of PDE solves
10−2
10−1
100tr
ueer
ror
||Q(u
(·,ξ)
)−
Q(U
(·,ξ)
)||L2 Ω
−12
−14
||ED||L2Ω
= O(10−3)
uniform h-refinementadaptive h-refinement
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 18 / 35
Numerical results
Example 3: Flow at low Reynolds numbers
Navier-Stokes equations:
−ν∆u+ u · ∇u+∇p = 0
∇ · u = 0
u = uin, x ∈ Γin
u = 0, x ∈ Γw ∪ Γcyl
σ · n = 0, x ∈ Γo
Parameterization of uncertainty: ν = ξ1
uin,x = ξ23
32(4− y2)
−2 −1 0 1 2 3 4 5 6−2
−1
0
1
2
Γw
Γw
Γi
Γo
Γcyl
QoI: Qξ (u) = ux (x0, ξ)
Let ξ1 ∼ U(0.01, 0.1), ξ2 ∼ U(1, 3)
s.t. Re =ξ2
8ξ1∈ [1.25, 37.5]
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 19 / 35
Numerical results
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 20 / 35
Numerical results
Example 3: Flow at low Reynolds numbers
104 105 106 107
Total Degrees of Freedom
10−4
10−3
10−2
10−1
||E|| L
2(Ξ
)
||Q|| L
2(Ξ
)
Ω and Ξ RefinementAdaptive RefinementΞ RefinementΩ Refinement
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 21 / 35
Efficient Bayesian inference - model selection for RANS
Adaptive Response Surface for Parameter Estimation
Objective
The main objective here is to develop a methodology based onresponse surface models and goal-oriented error estimation forefficient and reliable parameter estimation in turbulencemodeling.
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 22 / 35
Efficient Bayesian inference - model selection for RANS
Bayesian inference for RANS using surrogate models
Efficient Bayesian inference: Approximate response surface models canbe used to reduce the computational cost of the process.
• Ma and Zabaras, 2009.
• Li and Marzouk, 2014;Marzouk and Xiu, 2009;Marzouk and Najm, 2009.
UQ for RANS models: Uncertainty in the RANS model parameters is aknown issue in the turbulence community, but quantifying the effect of thisuncertainty is seldom analyzed in the computational fluid dynamicsliterature.
• Cheung et al., 2011.
• Oliver and Moser, 2011.
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 23 / 35
Efficient Bayesian inference - model selection for RANS
Fully developed incompressible channel flow
Mean flow equations: u = U + u′DUiDt
= −1
ρ
∂P
∂xi+
∂
∂xj
(ν∂Ui∂xj− u′iu′j
)∇ ·U = 0
Eddy viscosity assumption:
u′iu′j = −νT (Ui,j + Uj,i)
Channel equations: assuming homogeneous turbulence in x
∂
∂y
((ν + νT )
∂U
∂y
)= 1, y ∈ (0, H)
1Durbin and Petterson Reif, 2001; Pope, 2000S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 24 / 35
Efficient Bayesian inference - model selection for RANS
Model selection:• Set of modelsM = M1,M2, . . . ,Mn• Posterior plausibility = p(Mi|q,M)
• Likelihood =E(Mi|q,M) := p(q|Mi,M) =
∫Ξp(q|ξ,Mi,M)p(ξ|Mi,M) dξ
p(Mi|q,M) ∝ E(Mi|q,M) p(Mi|M)
1Calvetti and Somersalo, 2007; Jaynes, 2003; Kaipio and Somersalo, 2005S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 28 / 35
Efficient Bayesian inference - model selection for RANS
Calibration data:• Data is obtained from direct numerical simulation (DNS) 1
• Mean velocity measurements were taken at Reτ = 944 andReτ = 2003
Uncertainty models:• Three multiplicative error models
〈u〉+ (z; ξ) = (1 + ε(z; ξ))U+(z; ξ)
I independent homogeneous covarianceI correlated homogeneous covarianceI correlated inhomogeneous covariance
• Reynolds stress model⟨u′iu
′j
⟩+(z; ξ) = T+(z; ξ)− ε(z; ξ)
1Del Alamo et al., 2004; Hoyas and Jimenez, 2006S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 29 / 35
Efficient Bayesian inference - model selection for RANS
Numerical Results
Independent homogeneous covariance:
〈u〉+ (z; ξ) = (1 + ε(z; ξ))U+(z; ξ)⟨ε(z)ε(z′)
⟩= σ2δ(z − z′)
0.85 0.9 0.95 1 1.050
5
10
15
20
25
κ
0.7 0.8 0.9 1 1.10
5
10
15
cv1
0.8 1 1.2 1.4 1.6 1.80
1
2
3
4
5
σ
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 30 / 35
Efficient Bayesian inference - model selection for RANS
Numerical Results
Correlated homogeneous covariance:
〈u〉+ (z; ξ) = (1 + ε(z; ξ))U+(z; ξ)⟨ε(z)ε(z′)
⟩= σ2 exp
(−1/2
(z − z′)2
l2
)
0.85 0.9 0.95 1 1.050
5
10
15
20
κ
0.7 0.8 0.9 1 1.10
2
4
6
8
10
12
cv1
0.5 1 1.5 20
1
2
3
4
σ
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 31 / 35
Efficient Bayesian inference - model selection for RANS
Numerical Results
Correlated inhomogeneous covariance:
〈u〉+ (z; ξ) = (1 + ε(z; ξ))U+(z; ξ)⟨ε(z)ε(z′)
⟩= σ2
(2l(z)l(z′)
l2(z) + l2(z′)
)1/2
exp
(− (z − z′)2
l2(z) + l2(z′)
)
0.8 1 1.2 1.4 1.60
2
4
6
8
κ
0.8 1 1.2 1.4 1.60
1
2
3
4
cv1
0 2 4 6 80
0.5
1
1.5
σ
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 32 / 35
Efficient Bayesian inference - model selection for RANS
Numerical Results
Reynolds stress uncertainty:⟨u′iu
′j
⟩+(z; ξ) = T+(z; ξ)− ε(z; ξ)⟨
ε(z)ε(z′)⟩
= kin(z, z′) + kout(z, z′)
where kin models the error near the walls and kout far from the walls.
0.9 1 1.1 1.2 1.30
5
10
15
20
25
κ
0.8 1 1.2 1.4 1.60
2
4
6
8
cv1
0 2 4 6 80
0.5
1
1.5
σin
S. Prudhomme and C. Bryant Adaptive Surrogate Modeling January 6-9, 2015 33 / 35
Efficient Bayesian inference - model selection for RANS