Introduction to Finite Element Methods (FEM)

1. Introduction to Finite Element Methods (FEM)

 Definition: FEM is a numerical technique used to approximate solutions to boundary value

problems for partial differential equations (PDEs). It is widely used in engineering and
physics for solving complex problems.

 Need for FEM:

o Many real-world problems are too complex to solve analytically.

o FEM provides a way to break down complex geometries and material properties into
smaller, manageable parts.

 Background:

o Originated in the 1940s and 1950s for structural analysis.

o Evolved with advancements in computational power and software.

o Used in civil, mechanical, aerospace, biomedical, and electrical engineering.

2. Methods Employed in FEM - Steps in FEM

 Steps in FEM:

1. Problem Definition: Define the physical problem, including geometry, material

properties, and boundary conditions.

2. Discretization: Divide the domain into smaller, finite elements (meshing).

3. Element Equations: Derive equations for each element (e.g., stiffness matrix for
structural problems).

4. Assembly: Combine element equations to form a global system of equations.

5. Application of Boundary Conditions: Modify the global system to account for

constraints.

6. Solution: Solve the system of equations to find unknown variables (e.g.,

displacements, temperatures).

7. Post-Processing: Analyze and visualize results (e.g., stress, strain, heat flux).

3. Advantages, Disadvantages, Limitations, and Applications of FEM

 Advantages:

o Can handle complex geometries and boundary conditions.

o Applicable to a wide range of engineering problems.

 o Provides accurate results with proper meshing and modeling.

 Disadvantages:

o Computationally expensive for large-scale problems.

o Requires expertise in modeling and interpretation of results.

o Accuracy depends on mesh quality and assumptions.

 Limitations:

o Not suitable for problems with discontinuities or singularities without special

treatment.

o Requires significant computational resources for high-fidelity models.

 Applications:

o Structural analysis (stress, strain, deformation).

o Heat transfer, fluid dynamics, and electromagnetics.

o Biomechanics, geotechnical engineering, and acoustics.

 Concept of Discontinuity:

o Discontinuities (e.g., cracks, material interfaces) require special techniques like

enriched finite elements or adaptive meshing.

4. Phases of FEA (Finite Element Analysis)

1. Pre-Processing:

o Define the problem (geometry, material properties, loads, and boundary

conditions).

o Create a finite element mesh.

2. Solution:

o Solve the system of equations numerically.

3. Post-Processing:

o Analyse results (e.g., stress contours, deformation plots).

o Validate and interpret results.

5. Discretization Process

 Definition: The process of dividing a continuous domain into smaller, interconnected

subdomains (elements).

 Types of Elements:
 o 1D (bars, beams), 2D (triangles, quadrilaterals), 3D (tetrahedrons, hexahedrons).

 Considerations:

o Element size and shape affect accuracy and computational cost.

o Mesh refinement is often required in regions of high gradients.

6. Meshing – Element Type

 Meshing: The process of creating a finite element mesh.

 Element Types:

o 1D Elements: Used for trusses, beams, and frames.

o 2D Elements: Used for plane stress, plane strain, and axisymmetric problems.

o 3D Elements: Used for solid structures.

 Mesh Quality:

o Aspect ratio, skewness, and element distortion affect accuracy.

o Adaptive meshing can improve results in critical regions.

7. Stiffness Matrix of a Bar Element

 Bar Element: A 1D element used to model axial deformation.

 Stiffness Matrix:

o Represents the relationship between nodal displacements and forces.

o For a bar element with length L, cross-sectional area A, and Young’s modulus E:
1 −1
[k]=𝐸𝐴/𝐿
−1 1

8. Global Stiffness Matrix - Properties of Stiffness Matrix

 Global Stiffness Matrix:

o Assembled by combining stiffness matrices of individual elements.

o Represents the entire system’s response to applied loads.

 Properties:

o Symmetric: [K]=[K]T

o Positive Definite: Ensures stability of the solution.

o Sparse: Contains many zero entries, especially for large systems.

9. Boundary Conditions - Methods – Types

 Boundary Conditions (BCs):

o Essential BCs: Directly applied to nodal values (e.g., fixed displacement).

o Natural BCs: Applied to derivatives (e.g., applied forces).

 Methods of Applying BCs:

o Penalty Method: Introduces large stiffness values to enforce constraints.

o Lagrange Multipliers: Adds additional variables to enforce constraints.

 Types of BCs:

o Dirichlet BCs: Specify the value of the solution at boundaries.

o Neumann BCs: Specify the derivative of the solution at boundaries.

o Mixed BCs: Combination of Dirichlet and Neumann conditions.