Solving Elliptic (and Hyperbolic) Differential Equations in - - PowerPoint PPT Presentation

solving elliptic and hyperbolic differential equations in
SMART_READER_LITE
LIVE PREVIEW

Solving Elliptic (and Hyperbolic) Differential Equations in - - PowerPoint PPT Presentation

Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity Elasticity Hyperelasticity Viscoelasticity Bhavesh Shrimali Department of Civil and Environmental Engineering CS598APK, Fall 2017 Bhavesh


slide-1
SLIDE 1

Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity

Elasticity − → Hyperelasticity − → Viscoelasticity Bhavesh Shrimali

Department of Civil and Environmental Engineering

CS598APK, Fall 2017

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 1 / 12

slide-2
SLIDE 2

Introduction: Continuum Mechanics

Kinematics

Deformation mapping (χ) and Deformation Gradient (F) ∃χ (X) ∈ C 2 (Ω0) :    F = ∇χ ≡ Fij = ∂χi ∂Xj = ∂xi ∂Xj , 1 ≤ i, j ≤ 3 J = detF > 0 also u = χ − X = ⇒ F = I + ∇u It is difficult to analytically determine χ for most BVPs (Semi-inverse method, Fourier) or (FEM,BEM!)

Newton’s 2nd Law

Stresses (Cauchy and Piola-Kirchoff) ∃T : t = Tn &

b (x, t) dx +

  • ∂Ω

t (x, t) dx =

ρ (x, t) ¨ χ (x, t) dx

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 2 / 12

slide-3
SLIDE 3

More Continuum Mechanics...

  • v
  • v
  • e
  • X

x

  • AT = A

BT = B ∴ AB = BA Modeling Viscoelasticity – Two approaches

Hereditary Integrals: Stieltjes Integral S = JTF−T Internal variables (Increasingly popular!)

Two Potential Constitutive Framework: ψ and φ Constitutive Model:        S (F, Fv) = ∂ψ ∂F (F, Fv) ∂ψ ∂Fv + ∂φ ∂ ˙ Fv = 0 & DivS + B = 0

  • BLM

(1)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 3 / 12

slide-4
SLIDE 4

BVP

Isotropy and Non-negativity ψ (F, Fv) > 0 ψ (F, Fv) = ψ (QFK, Fv) ∀, Q, K ∈ U U =

  • A : AAT = ATA = I
  • Given a free energy function (ψ) and dissipation potential (φ), a domain (Ω0)

with smooth boundary (∂Ω0), choose an internal variable (Fv) and solve : DivS = 0 for X ∈ Ω0 (2) ∂ψ ∂Fv + ∂φ ∂ ˙ Fv = 0 at each time step (3) In general, the practice is to solve (3) at each time step (discretization) and then solve (2) using FEM

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 4 / 12

slide-5
SLIDE 5

Hyperelasticity (φ = 0)

For now, consider no dissipation and the following (ψ) (Convex!) ψ = µ 2 (I1 − 3) + κ 2 (J − 1)2 where I1 = F · F ≡ FijFij (Neo-Hookean) = ⇒ S = µF + κ (J − 1) JF−T ← −        ∂I1 ∂F = ∂ ∂F (F · F) = 2F ∂J ∂F = ∂ ∂F (detF) = JF−T

Underlying PDE

By balance of linear momentum, we finally get the PDE DivS = 0 = ⇒ µ∇ · F + κJ (J − 1) ∇ · F−T = 0 = ⇒ µ∇2u + κ∇ (J (J − 1)) F−T = 0 with

  • u

= g

  • n ∂Ωx

t = h

  • n ∂Ωt

(4) Equation (4) is the Cauchy-Navier equation for Hyperelasticity

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 5 / 12

slide-6
SLIDE 6

BVP: Set up

Quasi-static deformation of a spherical shell (R = |X|) For now, consider (J > 0), later we will consider (J = 1)

x

×10−3 −1.5 −1.0 −0.5 0.0 0.5 1.0 1.5

y

×10−3 −1.5−1.0−0.5 0.0 0.5 1.0 1.5

z

×10−3 −2.0 −1.5 −1.0 −0.5 0.0 0.5 1.0

Consider the domain on the left, given by Ω : X ∈ R3, A ≤ |X| ≤ B (5) where

  • A

= 10−3 B = 2 × 10−3 (6) Assumption: Radially symmetric deformation Points move radially outward Both Dirichilet and Neumann No bifurcations (Cavitation!)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 6 / 12

slide-7
SLIDE 7

Radially symmetric mappings

Consider deformation mapping (χ) of the form χ = f (R)X = ⇒ F = (Rf ′(R) + f ) 1 R2 X ⊗ X

  • K1

+f

  • I − 1

R2 X ⊗ X

  • K2

(7) = ⇒ F = λ1K1 + λ2K2 ⇐ ⇒ S = σ1K1 + σ2K2 (8)

Matrix Forms

The spectral forms of S and F S =   σ1 σ2 σ2   F =   λ1 λ2 λ2   (9) With some algebra, the BLM reduces to dσ1 dR + 2 R (σ1 − σ2) = 0 with f (A) = 1 , f (B) = 2 (10)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 7 / 12

slide-8
SLIDE 8

BVP... Finally

Therefore, the entire problem reduces to a single nonlinear ODE of the form 4

  • µ + κf 4

+ 2Rκf 3f ′2 + R

  • µ + κf 4

f ′′ = 0 (11) which reduces to f (R) + 2κ R

A

K(R, f ′(ξ), ξ) F (f (ξ)) dξ = G(R) (12) G(R) = 1 + c(R − A) − 4A R A

  • log

R A

  • − 1
  • + 1
  • (13)

K(R, f ′(ξ), ξ) = (R − ξ)f ′2(ξ) (14) F (f (ξ)) = f 3(ξ) µ + κf 4(ξ) (15) c = 1 + 4A B A

  • log

B A

  • − 1
  • + 1
  • + 2κ

B

A

K(B, f ′(ξ), ξ)F (f (ξ)) dξ (16)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 8 / 12

slide-9
SLIDE 9

Quadrature − → Nonlinear System

Using ideas from Linear IEs Nystr¨

  • m discretization

n-point Gauss-Legendre Quadrature to evaluate the integrals in (12) Nonlinear-Kernel fn(Ri) = Gn (Ri) − 2κ

N

  • j=1

ωjK (Ri, f ′

n(ξj), ξj) F (fn(ξj))

(17) Successive approximations f (k+1)

n

(Ri) = G(k)

n

(Ri) − 2κ

N

  • j=1

ωjK

  • Ri, f ′(k)

n

(ξj) , ξj

  • F
  • f (k)

n

(ξj)

  • (18)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 9 / 12

slide-10
SLIDE 10

Existence (and Uniqueness)

Nature of f ′(R)

Exact form of the kernel not reported in the literature For equations of the following form f (R) + R

A

K(R, ξ) ˆ F (f (ξ)) dξ = G(R) K(R, ξ) satisfies the Lipchitz condition ˆ F satisfies Lipchitz condition f (R) bounded and integrable G(R) bounded and integrable In general for nonlinear equations existence and uniqueness is not straightforward. Linearize (?) Do it for the linear problem

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 10 / 12

slide-11
SLIDE 11

Sample Results

0.0012 0.0014 0.0016 0.0018 0.0020 1000 2000 3000 4000

(a) f ′(R) vs R

0.0012 0.0014 0.0016 0.0018 0.0020 1.2 1.4 1.6 1.8 2.0

(b) f (R) vs R

Calculations from FEM

The gradient is sharp as R − → A+ Need more points to evaluate the integral (?)

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 11 / 12

slide-12
SLIDE 12

THANK YOU

Bhavesh Shrimali (UIUC) Solving Elliptic (and Hyperbolic) Differential Equations in Nonlinear Viscoelasticity CS598APK, Fall 2017 12 / 12