0% found this document useful (0 votes)
18 views

Slides (Aula1) : 2023/24 IST Semestre 1

Uploaded by

Raquel Ferreira
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
0% found this document useful (0 votes)
18 views

Slides (Aula1) : 2023/24 IST Semestre 1

Uploaded by

Raquel Ferreira
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/ 49

Slides (Aula1)

2023/24 IST Semestre 1

11Set2023
Sumário:
Apresentação: Docente, Programa, Bibliografia, Cronograma e Avaliação.
Introdução à Mecânica Computacional.
Modelos matemáticos discretos: molas, resistências electricas, …
Modelos matemáticos contínuos: barra elastica, transmissão de calor, …
Prob. 7a Formulação forte e formulação fraca simétrica.
Prob. 9 Problema de deformação elástica de barra
Prof. Miguel Matos Neves 2023/24 * DEM/IST Sem1
E-mail: [email protected]
Webpage: https://fenix.ist.utl.pt/homepage/ist13443

T4 Mecanica Computacional Aula **2asf 16h30-18h30 P12


Mecanica Computacional Aula **6asf 12h30-14h00 V1.12

T5Mecanica Computacional Aula **3asf 12h30-14h30 V1.24


Mecanica Computacional Aula **5asf 11h30-13h00 V1.14
Ver MeComAVA202324.pdf

Ficheiro: Slides_PFernandes_AulasT_MC_todas.pdf
Recommended Bibliography
• Introduction to the Finite Element Method: J.N Reddy 2006 McGraw-Hill
(The Reddy book is necessary for the open book exam).

• Respective Solutions Manual_JNReddy.pdf


A note on the chapters numbers in 3th
and 4th editions of the Reddy book
Practically the same with small changes in order and
others
======= 3th Edition ================== ============= 4th Edition ==================
Chapter numbers from Reddy book 3th Edition: Chapter numbers from Reddy book 4th Edition:
Introduction Chap. 1 Introduction Chap. 1
2nd ODEs in 1D (and 2D Trusses) Chap. 3, 4 2nd ODEs in 1D (Changed 2D Trusses) Chap. 3, 4
Scalar (Single-Variable) Bidimensional FEs Chap. 8, 9 --> Errors now in Chap. 3
Vectorial Bidimensional (Plane Elasticity) FEs Chap. 8, 11 --> Now 2D Trusses and Frames in new Chap. 6
4th order ODE in 1D (and 2D frames) Chap. 5 Scalar (Single-Variable) Bidimensional FEs Chap. 9, 8 e 10
Errors Chap. 14.5 Vectorial Bidimensional (Plane Elasticity) FEs Chap. 12
4th order ODE in 1D and 2D beams Chap. 5
Not strictly adopted, see lesson notes for:
Computer implementation in 1D Chap. 7 we use Matlab/Octave
Computer implementation in 2D Chap. 13 we use Matlab/Octave There is no chapter 14 in 4th Edition
===========================================
 Ver MeComAVA202324.pdf
Ver MeComAVA202324.pdf
Ver CronogramaMeC.Comp2023024.pdf

11set

18set

25set

02out Grupos

09out *5f,Enunc

16out
23out
30out
6nov
Ver CronogramaMeC.Comp2023024.pdf

13nov

20nov MAP

27nov *6f

4Dez *6f

11dez Entrega Relat 11dez

18dez
4jan
MAP(15jan?)
What about is Computational Mechanics?
• Computational mechanics is concerned with the application of computational
methods to study phenomena governed by the principles of mechanics.

• It is interdisciplinary and involves three pillars:


mechanics (fluid dynamics, thermodynamics, electromagnetics, solid mechanics,
etc),
mathematics, and
computer science.

• It was considered as a "third way" besides theoretical and experimental sciences (a


sub-discipline of applied mechanics). It is nowadays a sub-discipline within
computational science.
• Many physical phenomena can be described by differential equations.
These differential equations are not too difficult to deduce but can be too hard to
obtain their exact analytical solutions. In special for complex problem many times
it requires the use of approximation methods like the numerical methods (e.g.
Computational Mathematics).

• In this course we will learn how to obtain approximated solutions as near as


desirable/possible of the exact using the finite elemento method (FEM).

• The main idea with the FEM is to replace the continuous problem (e.g. a structure,
or a fluid) having infinite degrees of freedom (dofs) by an “equivalent” discrete
problem with a finite number of degrees of freedom.

• This discretization can be achieved by subdivision of the domain (a structure, or a


fluid) into several subdomains called finite elements.
Can you predict the displacements and stresses of this combined wrench? Sure.

3D mesh

https://www.comsol.com/blogs/parameterizing-dimensions-imported-cad-files/
Can we estimate
the mass and its
center?
(Complex Problem)

Wenzel, J., Sinapius, M. & Gabbert, U. CEAS Aeronaut J (2012) 3: 35. https://doi.org/10.1007/s13272-011-0040-6
Primary structure mass estimation in early phases of aircraft development using the finite element method
• Replacing a continuous domain with a collection of subdomains (“elements”) is not
new.
• There are many ways of calculating the circle area, perimeter or the number pi (up to
certain point), as well as ways of using polygons to do this. Archimedes used a
method of inscribing polygons and comparing their diameters to their perimeter to
get an estimation of pi. With more recent ways, you could use trigonometry or just
simple geometry.

Similar procedure from outside


Inscribing polygons will converge from upper values.
Convergence by lower
Many other discretizations can be
used, e.g. by rectangles, or
squares and triangles, etc.

https://sites.google.com/a/maret.org/advanced-math-7-final-project-2014/calculations/calculating-pi-by-inscribing-polygons
2016

M.J. Turner, Ray W. Clough, H.C.


Martin e L.J. Topp: Stiffness and
deflection analysis of complex
structures. Journal of the
Aeronautical Sciences, 23(9):805-
823, 1956

https://www.slideshare.net/MaryamShamsSolary/inventorstalk-62781058
The first paper on the finite element method (FEM) was published by Turner,
M. J., Clough, R. W., Martin H. C. and Topp, L. J. in 1956.
Clough coined the term “finite elements” in an article in 1960.
First book from Zienkiewicz
The Finite Element Method in Structural and
Continuum Mechanics, 1967
by Zienkiewicz & Cheung (Author)

Commercial codes as Abaqus, Ansys, and same


others appeared in 70’s.
The first commercial version of Ansys software
was labeled version 2.0 and released in 1971.
Abaqus FEA originally released in 1978.

Also from 70’s, Unigraph, later NX and NX


NASTRAN - Siemens NX (PrePost)
• Nowadays, with the improvements in computational power, it is possible to
model very complex systems and their surroundings using the FEM.

Contribution to the Fluid-Structure Interaction Analysis of Ultra-Lightweight Structures using an Embedded Approach –
Scientific Figure on ResearchGate. Available from:
https://www.researchgate.net/Octree-mesh-around-a-complex-airplane-model-The-picture-was-taken-from-Zudrop-in-48_fig25_275271188
[accessed 17 Sep, 2018]
Finite element analysis (FEA)
This numerical method used in many fields of
engineering and research. It allows efficient
and precise modelling the behavior of
mechanical, thermal, or other complex
systems.

This method separates a complex geometry


into a network of nodes and elements of
https://www.machinedesign.com/fea-and-simulation/what-s-difference-between-fem-fdm-and-fvm
simpler shape and equations, called a mesh.

http://www.truegrid.com/femgallery.html

https://www.lxsim.com/en/blog/what-are-the-main-applications-of-finite-element-analysis-fea/
• Stresses developed in screw fastener due to initial tightening and the service
load. The analysis involves contact and deformation.
Basic types of finite elements

Why not only one type of finite element for all applications?

http://feaforall.com/why-do-fea-engineers-use-1d-2d-3d-elements/
http://feaforall.com/why-do-fea-engineers-use-1d-2d-3d-elements/
http://feaforall.com/why-do-fea-engineers-use-1d-2d-3d-elements/
http://feaforall.com/why-do-fea-engineers-use-1d-2d-3d-elements/
Basic Steps of the Finite Element Method

A finite element analysis of a boundary-value problem (BVP) includes the following


basic steps:
1. Discretization or subdivision of the domain (Meshing).
2. Selection of the element i.e. interpolation functions (to provide an approximation
of the unknown solution within an element).
3. Assembling - Formulation of the system of equations ( also the major step in FEM.
The typical Ritz variational and Galerkin methods can be used.)
4. Solution of the system of equations
5. Postprocessing - once we have solved the system of equations, we can then
compute the desired parameters and display the result in form of curves, plots, or
color pictures, which are more meaningful and interpretable.)

1-3 is Preprocessing, 4 is solving and 5 is Postprocessing .


The first and the last step are the most relevant with the visualization process.

http://davis.wpi.edu/~matt/courses/fem/fem.htm
• 1D Discrete Models (Chapter 4): Springs
• 1D Discrete Models (Chapter 4): Electrical
Resistance
• 1D Continuous Models (Chapter 1)
Modelo Contínuo de Barra Elástica

Q: Como obtemos a equação diferencial? E as CFs?


Tal como vimos na última aula …
Modelo Contínuo de alheta com convecção

Q: Como obtemos a equação diferencial? E as CFs?


Tal como vimos na última aula …
(T-Tinf)

(T-Tinf)

(T-Tinf)

(T-Tinf)

CFs ???
Cont. Modelos matemáticos contínuos:
Cabo/catenária, torção, tubagens, … e analogias
To é tracção projectada
Hoje(P7a):
1.6 Problem 7

Consider the following problem

u( x)  u ( x)  1 0  x 1
u (0)  u (1)  0

to be solved by the finite element method.

a) Obtain the weak formulation of the problem.


b) Write the expressions for the stiffness matrix and the force vector and explain how they are obtained.
c) Establish the system of equations to be solved for a mesh of four linear elements.
d) Solve it and compare with the exact solution ( uexacto  0.269e x  0.735e x  1)
e) Solve the same problem using two quadratic elements. Compare the solution obtained with the
previous ones.
Hoje(P9):
1.9 Problema 9

Determine the reactions in the supports for the following problem:

E1 A1 P E2 A2

L1 L2

a) By classical methods.
b) By the finite element method.
Recapitulando:

Etapas FForte FFraca: Comentário final:


1. Definir o Resíduo r(x) Lembrar que nestes casos a matriz K é simétrica.
2. Método dos Resíduos Ponderados Int(r(x)*v(x) dx)=0 Assemblar de acordo com tabela de
3. Integração por partes Int[f’g]=Int[(fg)’]-Int[fg’] conectividade pode ser feito por blocos.
4. Introduzir as condições de fronteira Int[(fg)’]=[fg](a,b)

Etapas DORISP para da FFraca obter  a solução de Ku=F (Ax=b)


1. Discretizar (Malha)
2. Obter Eqs Elementares: Ke e Fe do element padrão Pré-processamento
3. Reunir Eqs Elementares: : tabela de conectividades + assembler Kg e Fg
4. Imposição das CFs Dirichelet, Neumann, …
5. Solução Ku=F Processamento
6. Pós-processar: Interpolar, derivar, gráficos, … Pós-processamento

* Próxima aula: P7b) +…

You might also like