method of finite elements i · method for . structural analysis • matrix method . for computing...

48
Institute of Structural Engineering Page 1 Method of Finite Elements I Chapter 2 The Direct Stiffness Method Method of Finite Elements I

Upload: others

Post on 21-Mar-2020

12 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 1

Method of Finite Elements I

Chapter 2

The Direct Stiffness Method

Method of Finite Elements I

Page 2: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 2

Method of Finite Elements I

Direct Stiffness Method (DSM)• Computational method for structural analysis• Matrix method for computing the member forces

and displacements in structures• DSM implementation is the basis of most commercial

and open-source finite element software• Based on the displacement method (classical hand

method for structural analysis)• Formulated in the 1950s by Turner at Boeing and

started a revolution in structural engineering

Page 3: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 3

Method of Finite Elements I

Learning Goals of this Chapter• DSM formulation• DSM software workflow for …

• linear static analysis (1st order)• 2nd order linear static analysis• linear stability analysis

Lecture 2b

Page 4: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 4

Method of Finite Elements I

Computational Structural Analysis

Modelling is the most important step in the process of a structural analysis !

X

Y

Physical problem Continuousmathematical model

Discretecomputational model

strong form weak form

Page 5: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 5

Method of Finite Elements I

System Identification (Modelling)

Global Coordinate SystemNodesElementsBoundary conditionsLoads

X

Y

1

3 4

25

6

1 2

3

4

Element numbersand orientation

Node numbers

5 6

Page 6: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 6

Method of Finite Elements I

Deformations – Degrees of Freedom

System Deformations

System identification

Nodal Displacements

nodes, elements, loads and supportsdeformed shape

(deformational, nodal)degrees of freedom = dofs

Page 7: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 7

Method of Finite Elements I

Degrees of Freedom

7 * 2 = 14 dof

Frame Structure

8 * 3 = 24 dof

Truss Structure

ui = ( udx , udy )

uiui

ui = ( udx , udy , urz )dof per node

dof of structure

Page 8: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 8

Method of Finite Elements I

Elements: Truss

compatibility

ux = displacement in directionof local axis X

ε =𝐷𝐷𝐷𝐷𝐿𝐿

σ = 𝐸𝐸 εconst. equation

equilibrium

𝑁𝑁 = 𝐴𝐴σ=𝐴𝐴𝐸𝐸 ε =𝐴𝐴𝐸𝐸𝐿𝐿𝐷𝐷𝐷𝐷

𝐹𝐹2 = −𝐹𝐹1 = 𝑁𝑁

𝐷𝐷𝐷𝐷 = (u2 −u1)

F2F1

𝐹𝐹1 =𝐴𝐴𝐸𝐸𝐿𝐿

(u1 −u2)

𝐹𝐹2 =𝐴𝐴𝐸𝐸𝐿𝐿

(−u1 + u2)

f = k uk : (element) stiffness matrixf : (element) nodal forcesu : (element) displacement vector

1 dof per nodeDX

𝐿𝐿,𝐸𝐸,𝐴𝐴N

F1 F2

ux

X/Y = local coordinate system

DX = displacement of truss end1 1

2 2

EA EAF uL LF uEA EA

L L

− =

Page 9: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 9

Method of Finite Elements I

Elements: Beam3 dof per node

DX

DY

RZ

𝐿𝐿,𝐸𝐸,𝐴𝐴uy

uy

ux

k u

ux = displacement in directionof local axis X

uy = displacement in directionof local axis Y

3 2 3 2

2

0 0 0 0

12 6 12 6 0 0

6 4 0 0

EA EAL L

EI EI EI EIL L L LEI EI

LL

2

3 2 3 2

2

6 2

0 0 0 0

12 6 12 6 0 0

6 2 0

EI EILL

EA EAL L

EI EI EI EIL L L LEI EI

LL

− − −

1

2

3

4

5

26

6 4 0

u

u

u

u

uEI EI

uLL

Page 10: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 10

Method of Finite Elements I

The Beam Stiffness Matrix

B

υA

υB

MAB

VAB

VBA

MBA

φB

φA

A

3 2 3 2

2 2

3 2 3 2

2 2

12 6 12 6

6 4 6 2

12 6 12 6

6 2 6 4

AB

AB

BA

BA

EI EI EI EIL L L L

V EI EI EI EIM L L L LV EI EI EI EI

L L L LMEI EI EI EIL L L L

− − = − − −

A

A

B

B

υϕυϕ

υA υBφA φB

4 DOFs φA, φB, υA, υΒ

Page 11: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 11

Method of Finite Elements I

υA

3 2 3 2

2 2

3 2 3 2

2 2

12 6 12 6

6 4 6 2

12 6 12 6

6 2 6 4

AB

AB

BA

BA

EI EI EI EIL L L L

V EI EI EI EIM L LL LV EI EI EI EI

L L L LMEI EI EI EI

L LL L

− − − − = − − − −

A

A

B

B

υϕυϕ

υB

* υΑ

* υB

The Beam Stiffness Matrix

Page 12: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 12

Method of Finite Elements I

* φA

φA

3 2 3 2

2 2

3 2 3 2

2 2

12 6 12 6

6 4 6 2

12 6 12 6

6 2 6 4

AB

AB

BA

BA

EI EI EI EIL L L L

V EI EI EI EIM L L L LV EI EI EI EI

L L L LMEI EI EI EIL L L L

− − = − − −

A

A

B

B

υϕυϕ

* φB BφBA

4EI L

26EL L2EI L

26EL L

φB

The Beam Stiffness Matrix

Page 13: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 13

Method of Finite Elements I

3 2 3 2

2 2

3 2 3 2

2 2

12 6 12 6

6 4 6 2

12 6 12 6

6 2 6 4

AB

AB

BA

BA

EI EI EI EIL L L L

V EI EI EI EIM L L L LV EI EI EI EI

L L L LMEI EI EI EIL L L L

− − = − − −

A

A

B

B

υϕυϕ

direct

: Moment/Shear in that occurs for a rotation/displacement / acting on node : Moment/Shear in that occurs for a rotation/displacement / acting on node

ik k k

ki i i

t i kt k i

ϕ δ

ϕ δ

Coupling Stiffness Terms

coupling

: Moment/Shear in that is required for inducing in a rotation/displacement / 1 : Moment/Shear in that is required for inducing in a rotation/displacement / 1

ik i i

ki k k

s i is k k

ϕ δ =

ϕ δ =

Direct Stiffness Terms

The Beam Stiffness Matrix

Page 14: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 14

Method of Finite Elements I

Elements: Global Orientation

local

global

uloc = R uglob

uglob = u = RT uloc

𝜃𝜃

kglob = k = RT kloc R

cos sin 0 0 0 0sin cos 0 0 0 00 0 1 0 0 00 0 0 cos sin 00 0 0 sin cos 00 0 0 0 0 1

R

θ θθ θ

θ θθ θ

= −

Page 15: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 15

Method of Finite Elements I

Beam Stiffness Matrixe.g. k24 =

reactionin global direction Yat start node S

due to a

unit displacementin global direction Xat end node E

UXE=1

FYS

S

EFXS =

FYS =

MS =

FXS =

FYS =

ME =

UXS UYS RS UXE UYE RE

k14 k15 k16

k24 k25 k26

k34 k35 k36

k44 k45 k46

k55 k56

k66

k11 k12 k13

k22 k23

k33

symm.

Element stiffness matrixin global orientation

[ ] [ ][ ] [ ]

iS iSSS SE

iE iEES EE

f uf u

=

i i

i i

k kk k

f = k u

Page 16: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 16

Method of Finite Elements I

Nodal Equilibrum

3 4

2

5

6

f4r4: Vector of all forces acting at node 4

r4 = - k6ES u3 + contribution of element 6 due tostart node displacement u3

- k6EE u4 + contribution of element 6 due toend node displacement u4

- k5EE u4 + contribution of element 5 due toend node displacement u4

- k5ES u2 + contribution of element 5 due tostart node displacement u2

f4 external load

Equilibrum at node 4: r4 = - k5SE u2 -k6ES u3 - k5EE u4 - k6EE u4 + f4 = 0

Page 17: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 17

Method of Finite Elements I

Global System of Equations

r1 = -

u1

r2 = -

r3 = -

r4 = -

u2 u3 u4

k5ES k6ES k5EE+k6EE

1

3 4

25

6

1 2

34

k1EE+k3SS+k4SS

k3SE k4SE

k3ES k2EE+k3EE+k5SS

k5SE

k4ES k4EE+k6SS

k6SE

+ f1 = 0

+ f2 = 0

+ f3 = 0

+ f4 = 0

- K U + F = 0 F = K U

Page 18: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 18

Method of Finite Elements I

K = global stiffness matrix = Assembly of all ke

F = K U

Global System of Equations

= equilibrium at every node of the structure

F = global load vector = Assembly of all fe

U = global displacement vector = unknown

Page 19: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 19

Method of Finite Elements I

Solving the Equation System

K U = F

U = K-1 F

What are the nodal displacements fora given structure (= stiffness matrix K )due to a given load (= load vector F ) ?

K-1left multiply

K-1 K U = K-1 F

Inversion possible only if K is non-singular(i.e. the structure is sufficiently supported = stable)

Page 20: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 20

Method of Finite Elements I

Beam Element Results

2. Element end forcesCalculate element end forces = f = k u

4. Element deformations along axis

1. Element nodal displacementsDisassemble u (indiv. element displacement vectors) from resulting global displacements U

3. Element stress and strain along axisCalculate moment/shear from end forces (equilibrium equation)Calculate curvature/axial strain from moments/axial force

Calculate displacements from strain (direct integration)

Page 21: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 22

Method of Finite Elements I

PP/2

PL/8 P/2

PL/8

l/2 l/2

P/2

PL/8P/2

PL/8

DSM Gleichung

Pfx1 fy2

1 2P/2

PL/8

P/2

PL/8

ux1

uy1

ux2

uy2

φ2φ1

intu + PK = ffx1

fy2

Special Considerations

Treatment of Loads acting within Elements

In this case, we need to transfer the loads within the element to equivalent nodal loads, and accordingly modify the DSM equation:

Page 22: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 23

Method of Finite Elements I

PP/2

PL/8 P/2

PL/8

l/2 l/2

P/2

PL/8P/2

PL/8

P/2

PL/8

P/2

PL/8

ux1

uy2

ux2

uy2

φ2φ1

= int+ Pf Ku

fx1fy2Lokale Elementkräfte

0

2

80

2

8

P

Pl

P

Pl

= −

intP

Special Considerations

Treatment of Loads acting within Elements

In this case, we need to transfer the loads within the element to equivalent nodal loads, and accordingly modify the DSM equation:

Page 23: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 24

Method of Finite Elements I

ql/2

qL2/12 ql/2l/2 l/2

ql/2

ql/2

ql/2 ql/2

qL2/12

ux1

uy2

ux2

uy2

φ2φ1fx1

fy2Lokale Elementkräfte

2

2

0

2

120

2

12

ql

ql

ql

ql

= −

intP

q

qL2/12

qL2/12 qL2/12

qL2/12

= int+ Pf Ku

Special Considerations

Treatment of Loads acting within Elements

In this case, we need to transfer the loads within the element to equivalent nodal loads, and accordingly modify the DSM equation:

Page 24: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 25

Method of Finite Elements I

- Discretization- Individualization/Localization- Member Formulation

- Global Coordinates- Assembly- Boundary Conditions- Solution for Displacements- Element Moment/Shears

The DSM-Steps

Page 25: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 26

Method of Finite Elements I

Illustrative Example: Truss System

Step # 0: Idealized System with Loads and Supports

1 2

3

fy3=1

fx3=2

Loads

Supports

The DSM-Steps

Page 26: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 27

Method of Finite Elements I@C. Felippa, Uni Colorado

Step #1: Define a global coordinate SystemEnumerate the Nodes, Elements and Degrees of Freedom

1 2

3

(3)(2)

(1)ux2, fx2

uy2, fy2

ux3, fx3

uy3, fy3

ux1, fx1

uy1, fy1

DOFs on the global system

x

y3

3 3

10 2

200 2

L

E A

=

=

2

2 2

1050

LE A

==

1

1 1

10100

LE A==

Illustrative Example: Truss System

The DSM-Steps

Page 27: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 28

Method of Finite Elements I@C. Felippa, Uni Colorado

Step #2: Decomposition and definition of Local Systems

1 2

3

(3) (2)

(1)

In this step we ignore external loads and supports and wedefine the local coordinate systems for every element

( ),i ix y

3x3y

2x

2y

1x1y

Illustrative Example: Truss System

The DSM-Steps

Page 28: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 29

Method of Finite Elements I@C. Felippa, Uni Colorado

Step #3: Definition of DOFs on the Local System

i j(e)

In this step, we define the DOFs for each element in their local system (according to each element type, e.g. beam or truss)

exey

, yyi iu f

, xxi iu f, yyj ju f

, xxj ju f

Illustrative Example: Truss System

The DSM-Steps

Page 29: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 30

Method of Finite Elements I

Step #4: Define the Local Stifness Matrix

For a Truss Element:

@C. Felippa, Uni Colorado

1 0 1 00 0 0 01 0 1 0

0 0 0 0

xixi

yiyi

xjxj

yjyj

ufuf EAuf Luf

− = −

1 0 1 00 0 0 01 0 1 0

0 0 0 0

eEAL

− = −

KTruss Stiffness Matrix:

xiu yiu xju yju

xiu yiu xju yju

xiu

yiu

xju

yju

Illustrative Example: Truss System

The DSM-Steps

Page 30: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 31

Method of Finite Elements I

Step #5: Global Coordinates – Rotation & Translation

@C. Felippa, Uni Coloradoi

j

exey

, yyi iu f

, xxi iu f

, yyj ju f , xxj ju f

xiu

yiu

xju

yju

( ) ( ), ,e ex y x y→

ϕ

cos sin 0 0sin cos 0 00 0 cos sin0 0 sin cos

xi xi

yi yi

xj xj

yj yj

u uu uu uu u

ϕ ϕϕ ϕ

ϕ ϕϕ ϕ

− =

& Te e e e e e= =

u R u u R u

Global

Local

Example: The truss element

In this step, we move from the local to the global coordinate system using the Rotation Matrix

The DSM-Steps

Page 31: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 32

Method of Finite Elements I

Step #5: Global Coordinates – Rotation & Translation

0 90

1 2

1 0 0 0 0 1 0 00 1 0 0 1 0 0 0

, 0 0 1 0 0 0 0 10 0 0 1 0 0 1 0

oϕ ϕ= =

− = =

R R

45

3

2 2 2 2 0 0

2 2 2 2 0 0

0 0 2 2 2 2

0 0 2 2 2 2

oϕ=

= −

R

For the elements of our simple example:

Example: The truss element

The DSM-Steps

Page 32: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 33

Method of Finite Elements I

Step #6: Transform the local stiffness matrix to global coordinates

@C. Felippa, Uni Colorado

1

e e e

e e e

Te e

e e e e e e e e

Te e e e e

Te e e e

==

=

= → =

→ = ⇒

=

u R uf R f

R R

K u f K R u R f

R K R u f

K R K R

Example: The truss element

The DSM-Steps

Page 33: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 34

Method of Finite Elements I

For our example:

@C. Felippa, Uni Colorado

1 2

3

(3) (2)

(1)

ux2, fx2

uy2, fy2

ux3, fx3

uy3, fy3

ux1, fx1

uy1, fy1

x

y

3

3 3

10 2

200 2

L

E A

=

=2

2 2

1050

LE A

==

1

1 1

10100

LE A==

Step #6: Transform the local stiffness matrix to global coordinates

Example: The truss element

The DSM-Steps

Page 34: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 35

Method of Finite Elements I@C.

Fel

ippa

, Uni

Col

orad

o

Example:

1 =T1 1 1K R K R

2 2 2 2= TK R K R

3 3 3 3= TK R K R

Step #6: Transform the local stiffness matrix to global coordinates

Example: The truss element

The DSM-Steps

Page 35: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 36

Method of Finite Elements I

DisassemblyIndividualization/LocalizationMember Formulation

Global CoordinatesAssemblyApplication of BCsSolutionResponse Quantities

Back to the Summary – DSM Steps

@C. Felippa, Uni Colorado

What have we done so far?

Next Steps:

Page 36: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 37

Method of Finite Elements I

Step #7: Assembly - Bring together the individual element stiffness matrices Keinto the System's stiffness Matrix Ksys

In our example we have 6 DOFs

1 =K

2 =K

3 =K

ux1 uy1 ux2 uy2

ux2 uy2 ux3 uy3

ux1 uy1 ux3 uy3

Stiffness Matrix remains Symmetrical!(Maxwell-Betti Prinzip)

U = [ux1 uy1 ux2 uy2 ux3 uy3]T

1 2 3 4 5 6

u1 u2 u3 u4

u3 u4 u5 u6

u1 u2 u5 u6

u1 u2u3u4

u3 u4u5u6

u1 u2u5u6

Example: The truss element

The DSM-Steps

Page 37: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 38

Method of Finite Elements I

1 1 1 111 21 31 411 1 1 121 22 32 42

1 1 1 1 131 32 33 431 1 1 141 42 43 44

k k k kk k k kk k k kk k k k

=

K

ux2 uy2 ux3 uy3

ux1 uy1 ux3 uy3

2 2 2 233 43 53 632 2 2 243 44 54 64

2 2 2 2 253 54 55 652 2 2 263 64 65 66

k k k kk k k kk k k kk k k k

=

K

3 3 3 311 21 51 613 3 3 321 22 52 62

3 3 3 3 351 52 55 653 3 3 361 62 65 66

k k k kk k k kk k k kk k k k

=

K

U = [ux1 uy1 ux2 uy2 ux3 uy3]T

1 2 3 4 5 6ux1 uy1 ux2 uy2

Step #7: Assembly - Bring together the individual element stiffness matrices Keinto the System's stiffness Matrix Ksys

In our example we have 6 DOFs

Example: The truss element

The DSM-Steps

Stiffness Matrix remains Symmetrical!(Maxwell-Betti Prinzip)

Page 38: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 39

Method of Finite Elements I

1 3 1 3 1 1 3 311 11 21 21 31 41 51 611 3 1 3 1 1 3 321 21 22 22 32 42 52 62

1 1 1 2 1 2 2 231 32 33 33 43 43 53 631 1 1 2 1 2 2 241 42 43 43 44 44 54 643 3 2 2 2 3 2 351 52 53 54 55 55 65 653 3 2 2 2 361 62 63 64 65 65

sys

k k k k k k k kk k k k k k k k

k k k k k k k kk k k k k k k kk k k k k k k kk k k k k k

+ ++ +

+ +=

+ ++ ++

K

2 366 66k k

+

ux1 uy1 ux2 uy2 ux3 uy3

1 2 3 4 5 6

U = [ux1 uy1 ux2 uy2 ux3 uy3]T

Stiffness Matrix remains Symmetrical! (Maxwell-Betti Prinzip)

Step #7: Assembly - Bring together the individual element stiffness matrices Keinto the System's stiffness Matrix Ksys

In our example we have 6 DOFs

Example: The truss element

The DSM-Steps

Page 39: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 40

Method of Finite Elements I

10 10 0 10 10 0 10 100 10 0 10 0 0 10 10

10 0 10 0 0 0 00 0 0 0 5 0 510 10 0 0 0 10 0 1010 10 0 5 0 10 5 10

sys

+ + − − − + + − − − +

= + −

− − + + − − − + +

K

ux1 uy1 ux2 uy2 ux3 uy3

where sys sys=K u f

Step #7: Assembly - Bring together the individual element stiffness matrices Keinto the System's stiffness Matrix Ksys

In our example we have 6 DOFs

Example: The truss element

The DSM-Steps

Stiffness Matrix remains Symmetrical! (Maxwell-Betti Prinzip)

Page 40: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 41

Method of Finite Elements I

1

1

2

2

21

x

y

xsys

y

RRR

fR

=

Step #8: Assemnle the nodal loads Fi into the System-Load Vektor Fsys

1 2

3

fy3=1

fx3=2

ux3

uy3

ux2

uy2ux1

uy1

Rx2

Ry2Ry1

Rx1

ux1

uy1

ux2

uy2

ux3

uy3

Example: The truss element

The DSM-Steps

Page 41: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 42

Method of Finite Elements I

Step #9: Impose the Boundar Conditions by eliminating the corresponding Rows and Columns of the fixed (restrained) DOFs from the system‘s Stiffness Matrix, and the system load vector.

ff fs f f

sys sys sys fs ss s s

= ⇒ =

K K u fK u f

K K u f

Why can we do this?

Denote with s the fixed DOFs and with f the fre DOFs, then theglbal equilibrium equations can be rewritten as:.

But for restrained DOFs we have: s =u 0

0

ff fs f f ff f f

sf ss s sf f s

== ⇒ =

K K u f K u fK K f K u f

This means we can solve the equation only for the free DOFs

Example: The truss element

The DSM-Steps

Page 42: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 43

Method of Finite Elements I

1

1

2

2

21

x

y

xsys

y

RRRR

=

f

1

1

2

2

3

3

x

y

xsys

y

x

y

uuuuuu

=

u

10 10 0 10 10 0 10 100 10 0 10 0 0 10 10

10 0 10 0 0 0 00 0 0 0 5 0 510 10 0 0 0 10 0 1010 10 0 5 0 10 5 10

sys

+ + − − − + + − − − +

= + −

− − + + − − − + +

K

ux1 = uy1 = ux2 = uy2 = 0

ffKfsK

sfKssK

Step #9: Impose the Boundar Conditions by eliminating the corresponding Rows and Columns of the fixed (restrained) DOFs from the system‘s Stiffness Matrix, and the system load vector.

Example: The truss element

The DSM-Steps

Page 43: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 44

Method of Finite Elements I

1

1

2

2

21

x

y

xsys

y

RRRR

=

f

1

1

2

2

3

3

x

y

xsys

y

x

y

uuuuuu

=

u

10 10 0 10 10 0 10 100 10 0 10 0 0 10 10

10 0 10 0 0 0 00 0 0 0 5 0 510 10 0 0 0 10 0 1010 10 0 5 0 10 5 10

sys

+ + − − − + + − − − +

= + −

− − + + − − − + +

K

ux1 = uy1 = ux2 = uy2 = 0

Step #9: Impose the Boundar Conditions by eliminating the corresponding Rows and Columns of the fixed (restrained) DOFs from the system‘s Stiffness Matrix, and the system load vector.

Example: The truss element

The DSM-Steps

Page 44: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 45

Method of Finite Elements I

21f

=

f

Step #10: Solution: Solve for the unknow displacements uff :

3

3

xf

y

uu

=

u 10 1010 15ff

= ⇒

K 1ff f f f ff f

−= ⇒ =K u f u K f

1 1für a b d bc d c aad bc

− − = ⇒ = −−

A AReminder: Inverse of a 2x2 Matrix:

Hence 1

3

3

15 10 2110 10 1150 100

0.40.2

f ff f

xf

y

uu

− − = = −−

⇒ = =

u K f

u

0000

0.40.2

sys

⇒ =

u

Example: The truss element

The DSM-Steps

Page 45: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 46

Method of Finite Elements I

Step #11: Back-Calculate the Individual Element forces(a) Reorganize the system (global) displacements usys into the respective elemental vectors ue

1 2 1

1 2 11 2 3

2 3 3

2 3 3

0 0 00 0 0

, , ,0 0.4 0.40 0.2 0.2

x x x

y y y

x x x

y y y

u u uu u uu u uu u u

= = = = = =

u u u

Example: The truss element

The DSM-Steps

Page 46: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 47

Method of Finite Elements I

Example: The truss element

The DSM-Steps

Step #11: Back Calculate the Element Forces(b) Rotation from global back to local coordinates

e e e=u R u

1 1 1 2 2 2

3 3 3

0 0 1 0 0 0 00 1 0 0 0 0 0

, 0 0 0 0 1 0.4 0.20 0 0 1 0 0.2 0.4

2 2 2 2 0 0 0 02 2 2 2 0 0 0 0

0.4 0.420 0 2 2 2 20.2 0.140 0 2 2 2 2

− = = = = =

− − − = = = − −

u R u u R u

u R u

Page 47: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 48

Method of Finite Elements I

Step #11: Back Calculate the Element Forces(b) Calculate the element forces

1 1 1

2 2 2

3 3 3

01 0 1 0 0 10 0 0 0 0 0

51 0 1 0 0.2 1

0 0 0 0 0.4 0

1 0 1 0 0 8.40 0 0 0 0 0

201 0 1 0 0.42 8.4

0 0 0 0 0.14 0

= =

− − = = = −

− − −

= = = −

F K u

F K u

F K u

eF

Example: The truss element

The DSM-Steps

Page 48: Method of Finite Elements I · method for . structural analysis • Matrix method . for computing the member forces and displacements in structures • DSM implementation is the basis

Institute of Structural Engineering Page 49

Method of Finite Elements I

Summary: Linear Static Analysis (1st order)

Workflow of computer program

1. System identification: Elements, nodes, support and loads2. Build element stiffness matrices and load vectors3. Assemble global stiffness matrix and load vector4. Solve global system of equations (=> displacements)5. Calculate element results

Exact solution for displacements and stresses