Sei sulla pagina 1di 44

# Swiss Federal Institute of Technology

Page 1

The Finite Element Method for the Analysis of Non-Linear and Dynamic Systems

Prof. Dr. Michael Havbro Faber Swiss Federal Institute of Technology ETH Zurich, Switzerland
Method of Finite Elements II

Page 2

## Contents of Today's Lecture

Motivation, overview and organization of the course Introduction to non-linear analysis Formulation of the continuum mechanics incremental equations of motion

Page 3

## Motivation, overview and organization of the course

Motivation In FEM 1 we learned about the steady state analysis of linear systems however, the systems we are dealing with in structural engineering are generally not steady state and also not linear We must be able to assess the need for a particular type of analysis and we must be able to perform it

Page 4

## Motivation, overview and organization of the course

Motivation What kind of problems are not steady state and linear? E.g. when the: material behaves non-linearly deformations become big (p- effects) loads vary fast compared to the eigenfrequencies of the structure General feature: Response becomes load path dependent

Page 5

## Motivation, overview and organization of the course

Motivation What is the added value of being able to assess the non-linear non-steady state response of structures ? E.g. assessing the: - structural response of structures to extreme events (rock-fall, earthquake, hurricanes) - performance (failures and deformations) of soils - verifying simple models

Page 6

Page 7

Page 8

Page 9

Page 10

Page 11

Page 12

## Motivation, overview and organization of the course

Steady state problems (Linear/Non-linear): The response of the system does not change over time

KU = R
Propagation problems (Linear/Non-linear): The response of the system changes over time

MU (t ) + CU(t ) + KU (t ) = R (t )
Eigenvalue problems: No unique solution to the response of the system

Av = Bv
Method of Finite Elements II

Page 13

Page 14

## Motivation, overview and organization of the course

Organization PowerPoint files with the presentations will be uploaded on our homepage one day in advance of the lectures http://www.ibk.ethz.ch/fa/education/FE_II

Page 15

Overview

Page 16

Overview

Page 17

Overview

Page 18

## Introduction to non-linear analysis

Previously we considered the solution of the following linear and static problem:

KU = R

## for these problems we have the convenient property of linearity, i.e:

KU = R, = 1 U = U, 1
If this is not the case we are dealing with a non-linear problem!

Page 19

## Introduction to non-linear analysis

Previously we considered the solution of the following linear and static problem:

KU = R
we assumed: small displacements when developing the stiffness matrix K and the load vector R, because we performed all integrations over the original element volume that the B matrix is constant independent of element displacements the stress-strain matrix C is constant boundary constraints are constant
Method of Finite Elements II

Page 20

## Classification of non-linear analyses

Type of analysis Description Typical formulation used Materiallynonlinear-only (MNO) Stress and strain measures used Engineering strain and stress

Materially-nonlinear Infinitesimal only displacements and strains; stress train relation is nonlinear Large Displacements and displacements, large rotations of fibers rotations but small are large; but fiber strains extensions and angle changes between fibers are small; stress strain relationship may be linear or non-linear Large Displacements and displacements, large rotations of fibers rotations and large are large; fiber strains extensions and angle changes between fibers may also be large; stress strain relationship may be linear or non-linear Method of Finite Elements II

Total Lagrange (TL) Second PiolaKirchoff stress, Green-Lagrange strain Updated Lagrange (UL) Cauchy stress, Almansi strain

Total Lagrange (TL) Second PiolaKirchoff stress, Green-Lagrange strain Updated Lagrange Cauchy stress, (UL) Logarithmic strain

Page 21

P 2

= P/ A = / E =L

E
1

P 2

< 0.04

Page 22

P 2 L

P/ A

Y = P/ A Y Y = +
E ET
1

ET

P 2

< 0.04

## Materially nonlinear only (infinitesimal displacements, but nonlinear stress-strain relation)

Method of Finite Elements II

Page 23

## Classification of non-linear analyses

y

< 0.04 = L

Large displacements and large rotations but small strains (linear or nonlinear material behavior)
Method of Finite Elements II

Page 24

## Classification of non-linear analyses

Large displacements, large rotations and large strains (linear or nonlinear material behavior)
Method of Finite Elements II

Page 25

P 2

P 2

Page 26

## Example: Simple bar structure

Area = 1cm 2
t

E = 107 N / cm 2 ET = 105 N / cm 2

u
t

Y
E
1

ET

Section a

Section b

La = 10cm
t

Lb = 5cm

R
4 3 2 1

Y = 0.002

Page 27

## Example: Simple bar structure

Area = 1cm 2
t

E = 107 N / cm 2 ET = 105 N / cm 2

ET

## Y : yield stress Y : yield strain

u
1
t

R
t

= 0.002
R
4 3 2

Section a

Section b

La = 10cm
t

Lb = 5cm

t u t u t a = , b = La Lb t

R + t b A = t a A

(elastic region)
t

= Y +

Y
ET

(plastic region)

Page 28

## Example: Simple bar structure

E = 107 N / cm 2 ET = 105 N / cm 2

t u t u a = , b = La Lb

Area = 1cm 2

u
t

Y
E

ET

## Y : yield stress Y : yield strain

R + t b A = t a A

R
t

(elastic region)
t

Section a

Section b

R
4 3 2 1

= 0.002

= Y +
=

Y
ET

(plastic region)

La = 10cm

La = 5cm

## Both sections elastic

t 1 1 R t R = EA t u ( + ) t u = La Lb 3 106

R 2 tR , b = a = 3A 3 A
Method of Finite Elements II

Page 29

## Example: Simple bar structure

t
E = 107 N / cm 2 ET = 105 N / cm 2

t u t u a = , b = La Lb

Area = 1cm 2

u
t

Y
E

ET

## Y : yield stress Y : yield strain

R + t b A = t a A

R
t

Section a

Section b

R
4 3 2 1

= 0.002

(elastic region)
t

La = 10cm

Lb = 5cm

= Y +
=

Y
ET

(plastic region)

## 3 section b will be plastic when R = Y A t t 2 u u a = E , b = ET ( Y ) Y La Lb

t

R
4 3 2 1
0.1 0.2
t

EA t u ET A t u t R= + ET Y A + Y A La Lb
t

u=

t

Page 30

## What did we learn from the example?

The basic problem in general nonlinear analysis is to find a state of equilibrium between externally applied loads and element nodal forces
t

R tF = 0 R = t R B + t R S + t RC F = t RI F=
m
t

We must achieve equilibrium for all time steps when incrementing the loading Very general approach includes implicitly also dynamic analysis!

t t

V (m)

B ( m )T t ( m ) t dV ( m )

Page 31

t +t

R t +t F = 0
t +t

assuming that
t +t

## R is independent of the deformations we have

F = tF + F

We know the solution tF at time t and F is the increment in the nodal point forces corresponding to an increment in the displacements and stresses from time t to time t+t this we can approximate by

F = t KU
Tangent stiffness matrix
Method of Finite Elements II

tF t K= t U

Page 32

## The basic approach in incremental analysis is

We may now substitute the tangent stiffness matrix into the equlibrium relation
t

KU =

t +t

R tF

t +t

U = tU+U

which gives us a scheme for the calculation of the displacements the exact displacements at time t+t correspond to the applied loads at t+t however we only determined these approximately as we used a tangent stiffness matrix thus we may have to iterate to find the solution
Method of Finite Elements II

Page 33

## The basic approach in incremental analysis is

We may use the Newton-Raphson iteration scheme to find the equlibrium within each load increment

t +t

K ( i 1) U ( i ) =

t +t

R t +t F ( i 1)

t +t

t +t

t +t

K (0) = t K ;

t +t

F (0) = t F

Page 34

## The basic approach in incremental analysis is

It may be expensive to calculate the tangent stiffness matrix and, in the Modified Newton-Raphson iteration scheme it is thus only calculated in the beginning of each new load step in the quasi-Newton iteration schemes the secant stiffness matrix is used instead of the tangent matrix

Page 35

## We look at the example again simple bar ( two load steps)

( t K a + t K b ) u ( i ) =
t +t

t +t

R ( t +t Fa ( i 1) t +t Fb ( i 1) )

t t t +t

t +t

Fa (0) = t Fa
t

t +t

Fb (0) = t Fb

CA ; Ka = La

Kb =

CA Lb

Page 36

## We look at the example again simple bar

Load step 1: t = 1: ( 0 K a + 0 K b )u (1) = 1R 1Fa(0) 1Fb(0) 2 104 u = = 6.6667 103 1 1 107 ( + ) 10 5 Iteration 1: (i = 1)
(1) 1 (1)

## = 1u (0) + u (1) = 6.6667 103 = =

1 (1)

1 (1) a

u = 6.6667 104 < Y (elastic section!) La u = 1.3333 103 < Y (elastic section!) Lb
1

1 (1) b 1

1 (1)

## Fa(1) = 6.6667 103 ;

Fb(1) = 1.3333 10 4

1

## ( 0 K a + 0 K b )u (2) = 1R 1Fa(1) 1Fb(1) = 0

Method of Finite Elements II

u = 6.6667 ` 3 10

Page 37

## We look at the example again simple bar

Load step 2: t = 2 : ( 1K a + 1K b ) u (1) = 2 R 2 Fa(0) 2 Fb(0) (4 104 ) (6.6667 103 ) (1.333 104 ) u = = 6.6667 10 3 1 1 107 ( + ) 10 5 Iteration 1: (i = 1)
(1) 2 2 2 1

## b(1) = 2.6667 103 > Y (plastic section!)

1 (1) Fb(1) = ( E T ( 2 b Y ) + Y ) A = 2.0067 10 4

Fa(1) = 1.3333 10 4 ;

## ( 1K a + 1K b ) u (2) = 2 R 2 Fa(1) 2 Fb(1) u (2) = 2.2 103

Method of Finite Elements II

Page 38

## We look at the example again simple bar

i 2 3 4 5 6 7

u (i)
1.45E-03 1.45E-03 9.58E-04 6.32E-04 4.17E-04 2.76E-04

(i)

Page 39

## The continuum mechanics incremental equations

The basic problem: We want to establish the solution using an incremental formulation The equilibrium must be established for the considered body in its current configuration

In proceeding we adopt a Lagrangian formulation where we track the movement of all particles of the body (located in a Cartesian coordinate system) Another approach would be an Eulerian formulation where the motion of material through a stationary control volume is considered
Method of Finite Elements II

Page 40

u1 u = u2 u 3

## Configuration corresponding to variation in displacements u at

t +t

Configuration at time t + t

x3

## Surface area Volume

t +t

t +t

Configuration at time t Surface area t S Configuration at time 0 Surface area 0 S Volume 0V Volume tV

x2
x1 (or 0 x1 , t x1 , t+t x1 )
Method of Finite Elements II

Page 41

## The continuum mechanics incremental equations

The Lagrangian formulation We express equilibrium of the body at time t+t using the principle of virtual displacements t +t t +t eij d t +tV = t +t R
V
x3

u1 u = u2 u 3

t +t

## Configuration at time t + t Surface area Volume

t +t t +t

Configuration at time t Surface area t S Configuration at time 0 Surface area 0 S Volume 0V Volume tV

x2
x1 (or 0 x1 , t x1 , t+t x1 )

t +t

t +t

t +t eij = (

## : Cartesian components of the Cauchy stress tensor u j 1 ui

2 t +t x j + t +t xi

t +t t +t t +t

## xi : Cartesian coordinate at time t + t R=

V : Volume at time t + t
t +t

t +t

f i B ui d t +tV +

t +t

Sf

t +t

fi S uiS d t +t S

Page 42

## The continuum mechanics incremental equations

The Lagrangian formulation We express equilibrium of the body at time t+t using the principle of virtual displacements
t +t
x3

u1 u = u2 u 3

## Configuration corresponding to variation in displacements u at

t +t

u
Configuration at time t + t Surface area Volume
t +t t +t

Configuration at time t Surface area t S Configuration at time 0 Surface area 0 S Volume 0V Volume tV

x2
x1 (or 0 x1 , t x1 , t+t x1 )

R=

t +t

t +t

fi B ui d t +tV +

t +t

Sf

t +t

fi S uiS d t +t S

where
t +t t +t t +t

fi B : externally applied forces per unit volume fi S : externally applied surface tractions per unit surface S f : surface at time t + t
t +t

Sf

Page 43

## The continuum mechanics incremental equations

The Lagrangian formulation We recognize that our derivations from linear finite element theory are unchanged but applied to the body in the configuration at time t+t

Page 44

## In the further we introduce an appropriate notation:

Coordinates and displacements are related as:
t

xi = 0 xi + t ui xi = 0 xi + t +t ui
t +t

t +t

t ui =

ui t ui

t +t S 0 i

t +t

+ ij = tt+tt ij

## Differentiation is indexed as:

t +t 0 i, j

t +t ui u = 0 , xj

0 t +t m , n

0 xm = t +t xn