Dr.

Mahmoud Eltaweel
Email: m.eltaweel@herts.ac.uk
7ENT2063-0105-FEM & Applications
Lecture 3: Introduction to Numerical Methods and
Discretization in Finite Element Analysis
 Learning Outcome 2
Intended learning Outcomes

At the end of this lecture, a successful student should be able to:

• LO1. Apply engineering principles for production of Computer-Aided Design (CAD) and Finite
Element Analysis (FEA) models presenting the required meshes and solutions settings for
complex engineering problems (M3)
 1. Introduction to Numerical Methods 3

1.1. What Are Numerical Methods?

• Many math problems cannot be solved analytically (exact solution).

• Numerical methods are approximate techniques.
• Real life problems in science and engineering require these numerical techniques.
• Real world problems can take hours, days, or years to solve. A well executed computer program can do it in
seconds.

Numerical solutions approximate behaviour at discrete points (nodes),

enabling analysis of intricate systems
 Analytical Analysis Numerical Analysis

Provide a direct solution and will Can be used with any function.
result in exact solution if one They often require many
exists iterations to get the true solution

Practical only for functions that Solution is usually not exact, and it
have a simple, closed-form is also necessary to provide initial
mathematical structure estimates of the unknowns

4
Numerical methods allow engineers to address:
• Complex geometries (e.g., irregular shapes in solid mechanics).
• Nonlinear material behaviour (e.g., non-isotropic properties).
• Boundary and loading conditions that are difficult to model analytically.

They enable the analysis of multi-physics problems like fluid-structure interaction and
heat transfer in conjunction with mechanical stresses.

5
 What is Finite Element Analysis? 6

The finite element method (FEM) is based on the idea of building a complicated object with simple blocks or
dividing a complicated object into smaller and manageable pieces.

Application of this simple idea can be found everywhere:

LEGO® toys:

Children play with LEGO® toys by using many small pieces, each of
very simple geometry, to build various objects.
With more and more smaller pieces, these objects will look more realistic.
 2. Introduction to the Finite Element Method (FEM) 7

2.1. What is FEM?

• FEM is a numerical method used to solve complex differential equations.
• It provides approximate solutions for problems that are difficult to solve analytically.
2.2. Basic Approach of FEM
• The problem domain is divided into smaller, manageable elements called finite elements.
• These elements are connected at specific points called nodes.
• This division simplifies the complex domain into discrete regions for easier analysis.
Continuous
System

Discrete
System

8
 2. Introduction to the Finite Element Method (FEM) 10
2.3. FEM Workflow
1. Geometry: Import or Create Geometry: Define the physical domain of the problem, such as a part, structure, or system.
2. Preprocess: Mesh and Define Physics:
• Generate a mesh to discretise the geometry into finite elements.
• Assign material properties, boundary conditions, and loading parameters to the model.
3. Solve: Solve Engineering Problems:
• Compute the solution using numerical methods to determine physical behaviours, such as stress, deformation, or
temperature.
4. Postprocess: Visualise and Analyse Results:
• Interpret the outcomes through visualisation tools to understand results such as contours, plots, and animations.
 3. Discretization in Finite Element Analysis 12
X
3.1. What is Discretization?
• The process of dividing a continuous domain into discrete elements.
• This allows complex systems to be broken down into smaller, manageable parts for numerical analysis.

3.2. Importance of Discretization

• Converts continuous partial differential equations into solvable algebraic equations.
 3. Discretization in Finite Element Analysis 13
X
3.3. Types of Elements
• 1D Elements: Used for structures like trusses and beams.
• 2D Elements: Applied to surfaces such as plates and shells.
• 3D Elements: For solid domains like mechanical components.
 3. Discretization in Finite Element Analysis 14

3.3. Concept of Mesh

Element Shapes: Triangles, quadrilaterals (2D), tetrahedrons, hexahedrons (3D).
Mesh Density: Fine mesh for higher accuracy, coarse mesh for faster computation.
Element Quality: Determines the reliability and convergence of the solution.
 3. Discretization in Finite Element Analysis 15

3.4. Balancing Accuracy and Computational Cost

• Accuracy: Fine meshes capture more detail but require higher computational resources.
• Cost: Coarser meshes reduce computation time but may sacrifice precision.

The goal is to achieve a balance based on the complexity of

the problem (Verification)
3. Discretization in Finite Element Analysis 16
 3. Discretization in Finite Element Analysis 17

Load–deflection curve for the reinforced concrete beam

analysed for three different meshes with crack band approach.
a) coarse mesh, (b) medium mesh, and (c) fine mesh.

Xenos, D. and Grassl, P., 2016. Modelling the failure of reinforced concrete with nonlocal and crack band approaches using the damage-plasticity model
CDPM2. Finite Elements in Analysis and Design, 117, pp.11-20.
 3.5. Types of Numerical Methods 18
Finite Difference Methods (FDM): X
• Based on replacing derivatives in differential equations with difference equations.
• Suitable for simpler problems but struggles with complex geometries and boundary conditions.
Finite Element Methods (FEM):
• Uses integral formulations to create algebraic equations.
• Divides the domain into smaller subregions (elements) and approximates the solution for each element.
 4. Practical Application in ANSYS Mechanical 19

4.1. What is ANSYS?

ANSYS is a leading FEA software used to solve engineering problems in areas such as structural analysis.

Setting Up a Simple FEA Problem in ANSYS

1. Geometry Creation: Define or import the physical model to analyse.

2. Material Properties Assignment: Assign appropriate material characteristics (e.g., Young's modulus,
thermal conductivity).
3. Mesh Generation and Refinement: Divide the geometry into elements for analysis.
4. Boundary Condition Application: Define loads (forces, pressures) and constraints (fixed, sliding).
5. Solving: ANSYS computes the solution by assembling and solving the governing equations.
6. Post-Processing: Visualise results such as stress, deformation, or temperature using contours, graphs, and
animations.
 Lecture Overview 20
Intended learning Outcomes

At the end of this lecture, a successful student should be able to:

• Understand numerical methods and their role in solving engineering problems.

• Explain the fundamentals of FEM and its workflow.
• Apply principles of discretisation, including mesh generation and refinement.
THANK
YOU
Mahmoud Eltaweel
m.eltaweel@herts.ac.uk

21