FEM Intro
FEM Intro
FEM Intro
Hey friends!
Welcome to Finite Element Analysis Convergence and Mesh Independence Study. This is a project-
based course which should take approximately 1 hour to finish. Before diving into the project,
please take a look at the course objectives and structure:
Resources
Updated Courses on Simulation, SimScale & More:
http://courses.jousefmurad.com/
First things first, here are some resources that are helpful for you during the class, make sure to
create a separate folder on your machine!
Mesh Convergence Study Spreadsheet (needed for this class - make sure to copy the spreadsheet!)
Mesh Convergence Presentation (from the class)
Mesh Convergence - Additional Exercises
The Finite-Element Method – A Beginner's Guide (check it out before or while you're doing the
course to grasp some concepts better)
Mesh Convergence – what is this?
First vs. Second Order Elements
Important Links
📥 Latest newsletter: https://jousef.substack.com/
Course Objectives
In this course, we are going to focus on two following learning objectives:
Course Structure
This course is divided into 3 parts:
1. Course Overview: This introductory reading material you are reading.
2. Finite Element Analysis Convergence and Mesh Independence Study: This is the hands-on project
that we will work on in Rhyme.
3. Graded Quiz: This is the final assignment that you need to pass in order to finish the course
successfully.
Project Structure
The hands on project on Finite Element Analysis Convergence and Mesh Independence Study is
divided into the following tasks:
Once you received your certificate, post in on your socials and tag me! One professional advice
from my side would be to add the certificate to your LinkedIn profile. Learn more about that here.
Let's connect and keep in touch for more upcoming Coursera guided projects!
LinkedIn
Twitter
Instagram
About Rhyme
This course runs on Coursera's hands-on platform called Rhyme. On Rhyme, you do projects in a
hands-on manner in your browser. You do no need any data, geometry/CAD files or any other
software, only a browser is required. So, you can just focus on the learning.
If you need help troubleshooting Rhyme, please refer to the Coursera Help Center for more
information.
Earn a Certificate
After you have completed the Finite Element Analysis Convergence and Mesh Independence Study
hands-on project, you will be able to assess your knowledge using an ungraded assignment. Once
you are comfortable with the concepts, take the final quiz, score higher than 80% to earn your
certificate.
The term Finite Element was introduced 1960 by Ray William Clough in his paper "The Finite Element
Method in Plane Stress Analysis". In the early 60s this method has been used by several engineers for stress
analysis, fluid transport, heat transport and other subjects
The first Finite-Element-Method book has been published by Olgierd Zienkiewicz, Richard Lawrence
Taylor and Jianzhong Zhu
In the late 60s and 70s the field of FEM application expanded and became a leading numerical approximation
in a broad field of engineering problems
Most commercial codes like ANSYS, ABAQUS, Adina and several others have their origin in the 1970s
John Swanson releases the first version of his so-called ANalysis SYStems (ANSYS) FEA software tool
(1970)
A web search (2006) for ‘finite element’ using Google yielded over 14 million pages compared to 2022 with
128 million pages
The number of papers have gone from one (1956) to more than a million (2022)
FEM systems at Mercedes in the early beginnings have been solved using punchcards (see video below)
Motivation 💪
Engineers use the finite element method for several good reasons. One of them being is that analytical models are
limited in their application. This is the case if we have:
complex geometries or deformation (e.g. crash testing)
complex loadings (e.g. time dependent force application)
complex material laws such as:
1. Anisotropic material law (fibre-reinforced plastic or crystalline material)
2. Inhomogeneous material properties (Young’s modulus…)
3. Hyperelastic material
and many more...
Numerical Methods 🧮
In general, there are three methods on how engineers are able to solve engineering problems.
1. Classical Methods
2. Numerical Methods
3. Experimental Methods
Whenever engineers solve complex problems involving complex geometries, loading conditions or material laws, they
cannot use classical analytical approaches using closed-form methods. Therefore numerical methods offer a way to
solve a problem where an analytical solution does not exist!
A closed form solution is an expression for an exact solution given with a finite amount of data.
The description of nature and the laws of physics for space- and time-dependent problems are usually expressed
with partial differential equations (PDEs). These equations are solved in an approximate manner by the finite
element method (FEM) which is based on equations of classical methods such as the Theory of Elasticity.
PDEs are equations for an unknown function of two or more independent variables that involves partial derivatives.
Below is an example of a PDE, namely the three-dimensional Laplace equation where�ϕis the dependent variable,
and x, y, and z are the spatial independent variables.
∂2�∂�2+∂2�∂�2+∂2�∂�2=0 ∂x2∂2ϕ+∂y2∂2ϕ+∂z2∂2ϕ=0
In order to solve this equation, it must be subjected to so-called initial conditions and boundary conditions.
An Initial Boundary Value Problem (IBVP) consists of the Partial Differential Equation the Initial Conditions as well
as the Boundary Conditions
Those BCs also known as Kinematic Boundary Conditions must be satisfied according to geometric constraints. For
example:
�(�=0)=0&�′(�=0)=0u(x=0)=0&u′(x=0)=0
at a fixed end of a cantilever.
This boundary condition can be a "fixed BC" in your FEA tool and constraints a shaft and its position in space
Boundary conditions which are related to the geometry, are called geometric boundary conditions
Those BCs also known as Static Boundary Conditions must be satisfied as a result of free and moment balances. For
example:
���2���2(�=�)=0EIδx2δ2u(x=L)=0
is a moment boundary condition at the free end of a cantilever. So, it is prescribed by forces and moments.
This is the boundary condition is prescribed as a load acting on your component
Boundary conditions which are related to loadings, are called force boundary conditions
Approach of the FEM 💻
From a mathematical point of view, the following five steps are essential to know in order to understand how the
FEM is working behind the scenes.
Step #1
Create geometry (CAD Model)
Define material properties
Choose initial and boundary conditions
Define other conditions such as contact behaviour
Pre-Processing, also called model preparation, is usually the most labor-intensive step of FEA.
Step #2
Element formulation 👉 development of equations for elements
Set up the partial differential equation in its weak form
Step #3
Assembly: Set up global problem 👉 obtaining equations for the entire system from the equations for one element
Step #4
Solving system of linear equations
Step #5
Post-Processing: determining quantities of interest, such as stresses and strains and obtaining visualisations of the
response
Hint: The solver may produce impressive, colourful results which may look convincing but can be completely useless.
Try to compare your solution with analytical results (if available), perform a proper convergence study and/or
compare your results with existing papers / literature.
There are obviously more points to check when validating and correlating your FEA results which will be covered in
another post. For now you only have to know not to trust your results and you have to employ methods to go through
a meticulous verification and validation process for every analysis.
In the section on FEA Workflow, we will have a look at a more granular level of how FEA works. In practise, you
are neither really busy thinking about how the matrix is assembled, nor do you care how the system of equations are
solved which is just a bunch of number crunching behind the scenes.
A characteristic feature of the finite element method is that instead of seeking the approximation over the entire
region, the region is divided into smaller parts, so called finite elements and the approximation is then carried out over
each element.
The collection of all small parts is called the finite elements.
When the type of approximation has been chosen (is to be applied over each element), the corresponding behaviour of
each element can then be determined. Having determined the behaviour of all single elements, the elements can then
be patched together (Matrix Assembly), which enables us to get an approximate solution of the entire body over the
entire domain.
As we mentioned before, we need to approximate a solution over our elements in order to determine the behaviour.
This approximation is usually a polynomial and is, in fact, some interpolation over the element. This means we know
some values at certain points within the element but not at every point. These ‘certain points’ are called nodal points
and are often located at the boundary of the element. The accuracy with which the variable changes is expressed by
the approximation, which can be linear, quadratic, cubic et cetera.
In order to get a better understanding of approximation techniques, we will look at a one-dimensional bar. Consider
the true temperature distribution T(x) along the bar in the picture below.
Equidistant grids can waste many nodes in areas where the solution is not important or changes in quantities
(gradients) are insignificant.
T
emperature Distribution Along a Rod
The nodal points we chose do not need to be equally spaced! We now divide our bar into four elements over which the
temperature is assumed to vary in a linear manner between each nodal point!
Divide & Conquer in Action - Dividing the Rod into Smaller Elements
Ele
mentwise Temperature Distribution
Temper
ature Distribution Using Linear Approximation
Well we see linear approximation is quite good, but we can certainly do better! If we chose a quadratic approximation
the temperature distribution along the bar is way more smooth.
Linear
vs. Quadratic Approach
Nevertheless, we see that irrespective of the polynomial degree the distribution over the fin is known once we know
the values at the nodal points, simple as that. If we just have one bar we would have an infinite amount of unknowns
(Degree of Freedom (DoF)). But in this case we have a problem with a finite number of unknowns.
A system with a finite number of unknowns is called a discrete system
A system with an infinite number of unknowns is called a continuous system
In our case we would have a system consisting of five equations which can be solved by hand but in general we have
thousands and thousands of unknowns. Luckily we have powerful machines at our fingertips allowing us to solve such
problems for us which makes life way easier.
FEA Workflow 💡
Part 1 – Geometry 📐
This is the part where you import your CAD model. Ideally, all redundant parts of the part that do not contribute to the
analysis or require excessive model setup have been removed
Many designers to create elaborate computerized models of objects before they are physically produced use
CAD modeling.
Step 1 -
Import your CAD Model
In general, there are two ways on how to create or import CAD models, depending on which FEA software you're
using.
Directly Import CAD: A very common solution is that your FEA tools allows you to import CAD models.
Obviously, this assumes that you already have a model available and modelled in your CAD software (Creo,
Solidworks etc.). There are tools such as ABAQUS that allow you to create your geometry inside the tool
itself.
Create CAD Inside the Tool: This is a method that can be used assuming that your tool of choice allows
creating a CAD model from scratch.
The definition of material and boundary conditions (BCs) can usually be captured at once .
Step 2 - Define
Material and Boundary Conditions
Material Definition: In this step, you have to define what material you will be using and define at the very
minimum the Young's modulus, the Poisson's ratio and material density. Once the basics are done, you can
even decide what material laws shall be applied such as linear elastic behaviour or if the material will have
plastic behaviour.
Material nonlinearity involves the nonlinear behaviours of material based on current deformation, temperature,
pressure and so on. Some nonlinear material models are large strain, stress-strain relationship, elastoplasticity,
plasticity, creep and hyperelasticity.
Boundary Conditions (BCs): This is one of the most complicated topics in FEA as you have to decide how
the real life conditions can be translated into a digital representation without simplifying it too much. Trust
me, this is way trickier than you think it is!
As a beginner, you can be quite sure that you will mess up boundary conditions very often as you not only have to
understand how BCs can be used in your model but also how the software is treating each BC.
Loads: Loads can be forces, moments, accelerations or even temperature loads on your components.
Contacts: Bodies can be connected to each other via common nodes. These connections are only possible if
the two bodies have the same mesh at the interface.
Contact is one type of nonlinearity of the system. An abrupt change in stiffness may occur when bodies come into or
out of contact with each other. This is a result of the changing nature of the contact between components in the
analysis during motion.
Meshing is an interesting but also difficult topic which has many aspects to consider. Personally, CAD cleanup is the
most tedious part whereas other engineers will say that meshing is the worst job for them - it's up to you to decide!
{�}=[�]{�}{F}=[K]{u}
which is the standard equation for the FEM where
{F} are the forces acting on the structure
[K] is the stiffness matrix
{u} are the displacements
You will see in another blog post, that if we only have a look at a linear spring element to derive the fundamental
equation for the FEM, the rows of the stiffness matrix are linearly dependent which means that the system
is statically underdetermined!
We have to integrate boundary conditions into the system, otherwise our solver will end up with an error.
Whenever we apply external forces and calculate displacements, this is called a Displacement-based procedure
Whenever we apply displacements and calculate the forces, this is called a Force-based procedure
Part 5 – Post-Processing 🎨
Post-Processing is the part where the numbers from the solver are taken and made interpretable for the engineers -
otherwise we would have a hard time decoding what all those numbers actually mean.
It is important to know it is not sufficient to simply accept the beautiful images that the machine throws at you. You
need to take your engineering knowledge and intuition and really think hard if the results actually make sense and are
physically accurate.
GIGO = Garbage In equals Garbage Out
Ste
p 5 - Post-Processing of the Results (v. Mises stress in MPa shown)
Part 6 – Iteration ↩️
Simple doing one analysis and looking at the plots will not do the job. FEA is an iterative process, which requires an
engineer to make adaptations to the model, fix boundary conditions and potentially perform several meshes to see if
there are no significant changes in stresses or other relevant output quantities the finer we discretise our domain.
If no convergence can be seen, this can be an issue of a so-called singularity.
A finite element model will sometimes contain so-called singularities, which means that there are points in your model
where values (such as stresses) tend towards an infinite value.
In the worst case, you have to go back to Step 1 and adapt your CAD model.
FEA Capabilities 🚀
FEA can be applied to different types of single or multiphysics problems involving heat transfer, fluid dynamics,
electric fields and more. Three main types of problems are:
Static: For example, structural analysis of different parts of a building or bridge when a certain load is applied
with no motion involved. Knowing what parts experience the highest stress tells the designers what parts need
to be strongest
Dynamic: Useful when the forces applied to the system change over time, for example, heat flow through a
system component
Modal: Useful in analysing the effects of vibration on a system
Stress Distribution of a Rack and Pinion Gear
Over the years, more "exotic" topics in FEA have emerged:
Extended FEM (XFEM): The word extended is added as the method enhances crack-propagation simulation
capability of the conventional finite elements formulation. This method is useful when analysing systems with
discontinuities, such as fractures; and answers questions how a system behaves when one or several
components have a crack or other defects.
Generalised FEM (GFEM): Combines traditional mesh-based FEM with more advanced "meshless"
methods. It combines the features of the traditional FEM and meshless methods. One of the advantages of
GFEM is the prevention of re-meshing around singularities.
Mixed FEM: Useful for problems involving contact or incompressibility, and where constraints are imposed
using Lagrange multipliers. These extra degrees of freedom arising from Lagrange multipliers are solved
independently. The system of equations is solved like a coupled system of equations.
hp-FEM: hp-FEM is a combination of automatic mesh refinement (h-refinement) and an increase in the order
of polynomial (p-refinement). It is however not the same as doing h- and p- refinements separately. The
method is used when certain elements in a system are further subdivided (in a process called mesh
refinement), and the polynomial equations of each element have different orders.
Discontinuous Galerkin FEM: DG-FEM has shown significant promise for utilising the idea of finite
elements to solve hyperbolic equations, where traditional finite element methods have been weak. Used to
analyse systems in which components bend and for incompressible problems which are typically observed in
most material processes.
Advantages of FEM 📈
1. Modelling of complex geometries and irregular shapes are easier as varieties of finite elements are available
for the discretisation of the domain
2. Different types of material properties can be easily accommodated, either from element-to-element or even
inside an element
3. Higher order elements may be implemented if the tools allow it
4. FEM is simple, well-defined and result-oriented and widely popular among an enthusiastic engineering
community that can help you whenever you run into problems
5. The availability of a large number of FEA software packages and literature makes FEM a versatile and
powerful, and accessible numerical method
Disadvantages of FEM 📉
Although FEM is an incredibly powerful method, there are however some drawbacks.
1. Simplification of real life components can’t be 100% exact for complex shapes
2. FEM is an approximate method and engineers are motivated to minimise the error over the whole domain
3. Although FEM is simple from its principles, the mathematics behind the FEM is relatively advanced, and thus
requires mathematical expertise for its implementation
4. The computation is very costly in FEM. As your mesh becomes finer, you will need more RAM. Almost all
FEM packages use the concept of bandwidth reduction to reduce the amount of memory needed to store the
stiffness matrix
5. The time needed for solving a FE problem increases with the degree of fineness of the mesh
6. Proper engineering judgement is needed to interpret results