Scribd
Upload
42 views9 pages

FINITE ELEMENT METHOD Lecture 5

The document discusses the finite element method for modeling problems in one dimension. It presents: 1) The general weak form and finite element formulation for a 1D problem, assuming linear basis functions and the Galerkin approach. 2) The resulting system of equations in matrix form, where the stiffness matrix K contains integral terms involving the basis functions. 3) Details of the linear 1D element, including derivation of terms in the stiffness matrix K and force vector f in local coordinates. 4) The algebraic system for a simple 1D problem modeled with linear elements. Extension to quadratic elements is also mentioned.

Uploaded by

Yasir Rasool
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
42 views9 pages

FINITE ELEMENT METHOD Lecture 5

The document discusses the finite element method for modeling problems in one dimension. It presents: 1) The general weak form and finite element formulation for a 1D problem, assuming linear basis functions and the Galerkin approach. 2) The resulting system of equations in matrix form, where the stiffness matrix K contains integral terms involving the basis functions. 3) Details of the linear 1D element, including derivation of terms in the stiffness matrix K and force vector f in local coordinates. 4) The algebraic system for a simple 1D problem modeled with linear elements. Extension to quadratic elements is also mentioned.

Uploaded by

Yasir Rasool
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

FINITE ELEMENT METHOD

BY

Prof.(Dr.) G.A.Harmain
Professor, Mechanical Engg. Dept.
National Institute of Technology
Hazrabal, Srinagar

Lecture 5
Finite Element Model in One dimension
 In general, Weak form can be written as:
𝑥𝑏 𝑑𝑊 𝑑𝑢
0= 𝑥𝑎
𝑎 + 𝑐𝑤𝑢 − 𝑊𝑓 𝑑𝑥 − (𝑤𝑄)𝑎 − (𝑤𝑄)𝑏
𝑑𝑥 𝑑𝑥

Assuming Linear solution 𝑢 ≈ 𝑢ℎ𝑒 𝑥 = ψ1 𝑢1 + ψ2 𝑢2


and 𝑊 = 𝑓 𝑢 and 𝑊𝑖 = ψ𝑖 (𝐺𝑎𝑙𝑒𝑟𝑘𝑖𝑛 𝐴𝑝𝑝𝑟𝑜𝑎𝑐ℎ)
When 𝑊1 = ψ1
𝑥𝑏
𝑑ψ1 𝑑ψ1 𝑑ψ2
0= 𝑎 𝑢1 + 𝑢2 + 𝑐ψ1 ψ1 𝑢1 + ψ2 𝑢2 − ψ1 𝑓 𝑑𝑥 − 1 × 𝑄1 − 0 × 𝑄2
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑑𝑥
𝑥𝑏 𝑥𝑏 𝑥𝑏
𝑑ψ1 𝑑ψ1 𝑑ψ1 𝑑ψ2
0= 𝑎 + 𝑐ψ1 ψ1 𝑢1 𝑑𝑥 + 𝑎 + 𝑐ψ1 ψ2 𝑢2 𝑑𝑥 − 𝑓 ψ1 𝑑𝑥 − 𝑄1
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎

𝑥𝑏 𝑑ψ1 𝑑ψ1 𝑥𝑏 𝑑ψ1 𝑑ψ2


0= 𝑥𝑎
𝑎 + 𝑐ψ1 ψ1 𝑢1 𝑑𝑥 + 𝑥𝑎
𝑎 + 𝑐ψ1 ψ2 𝑢2 𝑑𝑥 − 𝑓1 − 𝑄1
𝑑𝑥 𝑑𝑥 𝑑𝑥 𝑑𝑥
Finite Element Model in 1D (contd.)
When 𝑊2 = ψ2
𝑥𝑏
𝑑ψ2 𝑑ψ1 𝑑ψ2
0= 𝑎 𝑢1 + 𝑢2 + 𝑐ψ2 ψ1 𝑢1 + ψ2 𝑢2 − ψ2 𝑓 𝑑𝑥 − 0 × 𝑄1 − 1 × 𝑄2
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑑𝑥
𝑥𝑏 𝑥𝑏 𝑥𝑏
𝑑ψ2 𝑑ψ1 𝑑ψ2 𝑑ψ2
0= 𝑎 + 𝑐ψ2 ψ1 𝑢1 𝑑𝑥 + 𝑎 + 𝑐ψ2 ψ2 𝑢2 𝑑𝑥 − 𝑓 ψ2 𝑑𝑥 − 𝑄2
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎

𝑥𝑏 𝑑ψ2 𝑑ψ1 𝑥𝑏 𝑑ψ2 𝑑ψ2


0= 𝑥𝑎
𝑎 + 𝑐ψ2 ψ1 𝑢1 𝑑𝑥 + 𝑥𝑎
𝑎 + 𝑐ψ2 ψ2 𝑢2 𝑑𝑥 − 𝑓2 − 𝑄2
𝑑𝑥 𝑑𝑥 𝑑𝑥 𝑑𝑥

In matrix form:
𝑥𝑏 𝑥𝑏
𝑑ψ1 𝑑ψ1 𝑑ψ1 𝑑ψ2
𝑎 + 𝑐ψ1 ψ1 𝑑𝑥 𝑎 + 𝑐ψ1 ψ2 𝑑𝑥
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑢1 𝑓1 𝑄1
𝑢2 = +
𝑥𝑏
𝑑ψ2 𝑑ψ1 𝑥𝑏
𝑑ψ2 𝑑ψ2 𝑓2 𝑄2
𝑎 + 𝑐ψ2 ψ1 𝑑𝑥 𝑎 + 𝑐ψ2 ψ2 𝑑𝑥
𝑥𝑎 𝑑𝑥 𝑑𝑥 𝑥𝑎 𝑑𝑥 𝑑𝑥

𝐾11 𝐾12 𝑢1 𝑓1 𝑄1
𝐾21 𝐾22 𝑢2 = 𝑓2 + 𝑄2
Finite Element Model in 1D (contd.)
 In general, the 𝑖𝑡ℎ algebraic equation of the system of n equations
can be written as:
𝑛

𝐾𝑖𝑗 𝑢𝑗 − 𝑓𝑖 − 𝑄𝑖 𝑖 = 1,2, … 𝑛
𝑗=1
𝑥𝑏 𝑑ψ𝑖 𝑑ψ𝑗 𝑥𝑏
where 𝐾𝑖𝑗 = 𝑥𝑎
𝑎 + 𝑐ψ𝑖 ψ𝑗 𝑑𝑥 and 𝑓𝑖 = 𝑥𝑎
𝑓 ψ𝑖 𝑑𝑥
𝑑𝑥 𝑑𝑥
Thus
𝐾11 𝑢1 + 𝐾12 𝑢2 + ⋯ + 𝐾1𝑛 𝑢𝑛 = 𝑓1 − 𝑄1
𝐾21 𝑢1 + 𝐾22 𝑢2 + ⋯ + 𝐾2𝑛 𝑢𝑛 = 𝑓2 − 𝑄2
⋮
𝐾𝑛1 𝑢1 + 𝐾𝑛2 𝑢2 + ⋯ + 𝐾𝑛𝑛 𝑢𝑛 = 𝑓𝑛 − 𝑄𝑛
Finite Element Model in 1D (contd.)
In matrix form:
𝐾11 𝐾12 𝐾1𝑛 𝑢1 𝑓1 𝑄1
⋯ 𝑢2
𝐾21 𝐾22 𝐾2𝑛 𝑓2 𝑄2
⋮ ⋱ ⋮ ⋮ = ⋮ + ⋮
𝐾𝑛1 𝐾𝑛2 ⋯ 𝐾𝑛𝑛 𝑢𝑛 𝑓𝑛 𝑄𝑛
𝐾 𝑢 = 𝑓 + 𝑄

Where
Matrix K is called coefficient matrix or stiffness matrix,
Column vector f is source vector or force vector,
u are called primary element nodal degrees of freedom,
Q is secondary element nodal degrees of freedom
Linear Element
 For a mesh of linear elements, the element Ω𝑒 is located
between the global nodes 𝑥𝑎 = 𝑥𝑒 and 𝑥𝑏 = 𝑥𝑒+1
𝑒
𝑥𝑒+1 𝑑ψ𝑒𝑖 𝑑ψ𝑗
Hence 𝐾𝑖𝑗𝑒 = 𝑥𝑒
𝑎𝑒 + 𝐶𝑒 ψ𝑒𝑖 ψ𝑒𝑗 𝑑𝑥 and
𝑑𝑥 𝑑𝑥

𝑥𝑒+1
𝑓𝑖𝑒 = 𝑥𝑒
𝑓𝑒 ψ𝑒𝑖 𝑑𝑥

In local co-ordinate system 𝑥,


ℎ𝑒 𝑑ψ𝑒𝑖 𝑑ψ𝑒𝑗
𝐾𝑖𝑗𝑒 = 𝑎𝑒 + 𝐶𝑒 ψ𝑒𝑖 ψ𝑒𝑗 𝑑𝑥 and
0 𝑑𝑥 𝑑𝑥
ℎ𝑒
𝑓𝑖𝑒 = 0
𝑓𝑒 ψ𝑒𝑖 𝑑 𝑥
where x = 𝑥𝑖𝑒 + 𝑥 and
𝑑ψ𝑒𝑖 𝑑ψ𝑒𝑖
dx = d𝑥, =
𝑑𝑥 𝑑𝑥
Linear Element
𝑥
We have ψ1𝑒 𝑥 =1− ,
ℎ𝑒

𝑥
ψ𝑒2 𝑥 =
ℎ𝑒

ℎ𝑒 𝑑ψ1 𝑑ψ1
Hence 𝐾11 = 0
𝑎 + 𝑐ψ1 ψ1 𝑑𝑥
𝑑𝑥 𝑑𝑥
ℎ𝑒
1 1 𝑥 𝑥
= 𝑎 − − +𝑐 1− 1− 𝑑𝑥
0 ℎ𝑒 ℎ𝑒 ℎ𝑒 ℎ𝑒

𝑎 1
𝐾11 = + 𝑐ℎ𝑒
ℎ𝑒 3

Similarly derive 𝐾12 , 𝐾21 and 𝐾22


Linear Element (contd.)
ℎ𝑒 ℎ𝑒 𝑥 1
 Also 𝑓1𝑒 = 0
𝑓𝑒 ψ𝑒𝑖 𝑑𝑥 = 0
𝑓𝑒 1 − 𝑑𝑥 = 𝑓ℎ𝑒
ℎ𝑒 2

and calculate 𝑓2𝑒


 Hence linear algebraic equation form will be
𝑎 1 𝑎 1 𝑓ℎ𝑒
+ 𝑐ℎ𝑒 − + 𝑐ℎ𝑒
ℎ𝑒 3 ℎ𝑒 6 𝑢1 2 𝑄1
𝑢 = +
𝑎 1 𝑎 1 2 𝑓ℎ𝑒 𝑄2
− + 𝑐ℎ𝑒 + 𝑐ℎ𝑒
ℎ𝑒 6 ℎ𝑒 3 2

𝑎 1 −1 𝑐ℎ𝑒 2 1 𝑢1 𝑓ℎ𝑒 1 𝑄1
+ 𝑢2 = 2 1 + 𝑄2
ℎ𝑒 −1 1 6 1 2
 Similarly derive for Quadratic Element
7 −8 1 4 2 −1 𝑢1 1 𝑄1
𝑎 𝑐ℎ𝑒
−8 16 −8 + 30 2 16 2 𝑢2 = 𝑓ℎ𝑒 4 + 𝑄2
3ℎ𝑒 6
1 −8 −7 −1 2 4 𝑢3 1 𝑄3

You might also like