CT 7

FEM + shape functions + eigenvectors + modes

Summary of Key Concepts in Finite Element Analysis (FEM)

Shape Functions:
Definition: Mathematical functions used to interpolate field variables (like displacement) across the
element based on nodal values.

E.g.:

Properties:

Continuity: Ensures smooth transitions between elements.

Linear Independence: Guarantees that each shape function contributes unique information to the
solution.

Partition of Unity: The sum of all shape function values equals one at any point within an element.

Compact Support: Shape functions are non-zero only within a limited region, allowing for sparse
matrices and improved computational efficiency.

Types of Shape Functions:

1D Linear Elements: Shape functions increase linearly from 0 to 1 at their respective nodes.

2D Quadrilateral Elements: Shape functions decrease bilinearly from 1 at one corner to 0 at others.

3D Hexahedral Elements: Shape functions decrease trilinearly, indicating behavior at a vertex while
contributing to the overall element representation.
Graph descriptions:

1D Linear Finite Element Shape Function N1(ξ)N_1(ξ)N1(ξ): This plot illustrates how the
shape function associated with the node at ξ=−1ξ = -1ξ=−1 varies from 1 at that node to 0 at
the other node ξ=1ξ = 1ξ=1.

2D Quadrilateral Shape Function N3(ξ,η)N_3(ξ, η)N3(ξ,η): The 3D surface plot shows the
shape function associated with the corner at (ξ=1,η=1)(ξ = 1, η = 1)(ξ=1,η=1). It equals 1 at
that corner and decreases bilinearly to 0 at the other corners.

3D Hexahedral Shape Function NNN: The scatter plot demonstrates the shape function for
the vertex at (ξ=−1,η=−1,ζ=1)(ξ = -1, η = -1, ζ = 1)(ξ=−1,η=−1,ζ=1) in a hexahedral element. It
is equal to 1 at that vertex and decreases trilinearly to 0 at all other vertices.

Isoparametric Formulation:

Utilizes the same shape functions for both geometry and the solution field, simplifying the mapping
between local and global coordinate systems.

Interpolation Property:

Each shape function is equal to 1 at its own node and 0 at all other nodes, ensuring accurate
representation of nodal values.

Rigid Body Motion:

The partition of unity allows the finite element method to represent rigid body motion without
introducing strain, crucial for accurately modeling physical systems.

Finite Element Mesh:

The arrangement of elements (1D, 2D, 3D) impacts the analysis, and the choice of shape functions is
integral to the overall solution accuracy.

Applications of FEM:
Used to solve complex problems in solid mechanics, such as stress analysis, deformation, and
dynamic response of structures under various loads.

HOW DO SHAPE FUNCTIONS, EIGENVECTORS AND MODES LINK?

Definitions

1. Shape Functions:

o Definition: Shape functions are mathematical functions used in the finite element
method to interpolate the values of a field variable (such as displacement) across an
element based on the values at its nodes.

o Characteristics: Each shape function is associated with a node and has the property
that it is equal to 1 at its own node and 0 at all other nodes. This allows for the
representation of how the displacement or other physical quantities vary across the
element.

2. Strain Energy:

o Definition: Strain energy is the energy stored in a material due to deformation (e.g.,
stretching, compressing, or bending). It is a measure of the work done on the
material to deform it.

o Mathematical Expression: In the context of FEM, the strain energy UUU for a linear
elastic material can be expressed as:

where σ is the stress, ϵ is the strain, and V is the volume of the element. Strain
energy is directly related to the stiffness of the material and the shape of the
element.

3. Mode Shapes:

o Definition: Mode shapes are the shapes that a structure assumes when it vibrates at
specific natural frequencies. Each mode shape corresponds to a specific vibration
mode of the structure.

o Characteristics: In FEM, mode shapes can be represented as vectors of nodal

displacements that describe the deformation of the structure during free vibration.
 4. Eigenvalues:

o Definition: Eigenvalues are scalars associated with a linear transformation that

provide information about the behaviour of dynamic systems. In the context of FEM,
eigenvalues correspond to the natural frequencies of a structure.

o Mathematical Relation: Eigenvalues can be determined by solving the characteristic

equation of the system:

where [K] is the stiffness matrix, [M] is the mass matrix, λ represents the eigenvalue
(squared natural frequency), and [ϕ] (phi) represents the mode shape.

Relationships Between Concepts

 Shape Functions and Strain Energy: Shape functions are essential for calculating the strain
energy in finite elements. The displacement field across an element is expressed using shape
functions, which allows the computation of strain from displacements. Once strains are
computed, they can be used to determine stresses, and subsequently, the strain energy can
be calculated.

 Strain Energy and Eigenvalues: In dynamic analysis, the total strain energy associated with
the deformation of a structure can be linked to its eigenvalues. The natural frequencies of
vibration (which correspond to the eigenvalues) depend on the stiffness of the structure
(related to strain energy) and its mass distribution.

 Mode Shapes and Eigenvalues: Each eigenvalue corresponds to a mode shape. The mode
shape describes the deformation pattern of the structure at a specific natural frequency
(eigenvalue). Understanding the relationship between mode shapes and eigenvalues is
crucial for predicting how structures will respond to dynamic loading.

In summary, shape functions, strain energy, mode shapes, and eigenvalues are interconnected
concepts in finite element analysis. Shape functions are used to interpolate displacements, which
enables the calculation of strain energy. Strain energy influences the dynamic behavior of structures,
which is characterized by eigenvalues and mode shapes. The example of a vibrating cantilever beam
illustrates how these concepts work together to analyze structural behavior in FEM.
Connecting Eigenvalues, Strain Energy, and Stiffness

 Dynamic Response: When a structure vibrates, it deforms and stores strain energy. The
amount of energy stored is directly related to the stiffness of the structure. A stiffer
structure will store more energy for the same displacement compared to a less stiff
structure.

 Influence on Natural Frequencies: The eigenvalues (natural frequencies) of the system are
influenced by the stiffness matrix. A higher stiffness typically results in higher eigenvalues
(natural frequencies), meaning that the structure will vibrate at higher frequencies.

 Stiffness and Damping: In dynamic systems, damping also plays a role. While stiffness
dictates the natural frequencies, damping will influence the energy dissipation and the
overall dynamic response. However, it is the stiffness that primarily determines the strain
energy storage during vibrations.

Shape functions in FEM:

1. Element Definition
  In FEM, the physical domain (like a beam, plate, or solid object) is divided into smaller,
simpler parts called elements (e.g., 1D line elements, 2D triangular or quadrilateral
elements, and 3D tetrahedral or hexahedral elements).

 Each element has nodes at its corners, where the values of the field variable (like
displacement, temperature, or pressure) are defined.

2. Interpolation of Field Variables

 Shape functions are used to interpolate the values of the field variable across the entire
element based on the values at the nodes.

 For example, in a 1D element with nodes at ξ=−1 and ξ=1, a shape function N1(ξ) is defined
such that:

o N1(−1)=1 (value at the left node),

o N1(1)=0 (value at the right node).

 The value of the field variable u at any point ξ within the element can be calculated as:

where u1 and u2 are the values at the nodes.

3. Deriving Element Equations

Element-Level Calculations:

 For each element, use shape functions to calculate local stiffness matrices, strain-
displacement matrices, and mass matrices.

 Formulate the local strain energy and derive nodal displacements using the local stiffness
matrix.

 Shape functions are utilized to derive the element stiffness matrix and load vector. These
matrices relate the nodal displacements to the internal forces or stresses within the
element.
  For linear elastic problems, this involves using the shape functions to express the strain in
terms of the nodal displacements:

 [The strain energy for the element can then be derived and integrated over the element's
volume or area to form the element stiffness matrix.]

4. Assembly of Global System

 Each element contributes to the global system of equations. The local stiffness matrices
(derived using shape functions) are assembled into a global stiffness matrix.

 This involves mapping the local element contributions to the global node numbering,
ensuring that interactions between elements are captured correctly.

 After calculating the element stiffness matrices for all elements, they are assembled into a
global stiffness matrix [K] for the entire structure. The assembly process ensures that
contributions from each element are correctly included in the overall system.

5. Applying Boundary Conditions

 Once the global system of equations is constructed, boundary conditions are applied to the
nodal values. This can include fixed supports, applied forces, or displacements at specific
nodes.

6. Solving the System of Equations

 The resulting system of equations, often represented as (where [K] is the

global stiffness matrix, [u] is the displacement vector, and [F] is the force vector), is solved
using numerical methods (like Gaussian elimination or iterative solvers).

 The solution provides the nodal displacements.

7. Post-Processing

 Once the nodal displacements are obtained, they can be used to calculate stresses and
strains within each element.

 This is done by substituting the nodal displacements back into the shape functions to find
the values of the field variable at any point in the element.
CT 8

Strain Energy in FEM

1. Strain Energy:

o Definition: The potential energy stored in a material due to deformation under load.

o Significance: It is associated with eigenvalues and mode shapes, indicating how

energy is distributed in different deformation modes.

2. Mode Shapes and Eigenvalues:

o Mode Shapes: Characteristic deformation patterns of a structure when it vibrates.

o Eigenvalues: Associated with mode shapes, indicating natural frequencies of

vibration and the strain energy related to those modes. They reflect the stiffness
characteristics of the structure.

3. Global Stiffness Matrix Assembly:

o Process: Involves combining individual element matrices into a global matrix,

ensuring that contributions from shared nodes are appropriately accounted for by
adding stiffness contributions at those nodes.

4. Consistent Nodal Loads:

o Defined as equivalent nodal forces derived from distributed loads, ensuring accurate
representation in the analysis.

5. Higher Modes:

o Generally require more energy to deform and contribute less to the overall
structural response. Lower modes are usually more significant for structural
behavior under typical loading conditions.

6. Zero Energy Modes:

o Indicate rigid body motions that do not change the internal energy of the structure,
important for understanding dynamic behavior and constraints in the system.

7. Degrees of Freedom:

o The number of independent displacements or rotations a node can undergo,

affecting the number of observable mode shapes in a structure.

8. Eigenvalue Significance:

o In addition to representing natural frequencies and stiffness, they also define the
strain energy associated with each mode shape, crucial for evaluating how different
modes contribute to structural behavior.

Study Tips:
 Understand Relationships: Focus on how strain energy relates to eigenvalues and mode
shapes, and how these concepts tie into overall structural behavior.

 Matrix Assembly: Practice assembling global stiffness matrices from individual elements and
understand the implications of shared nodes.

 Dynamic Behaviour: Study the significance of higher and lower modes in dynamic loading
scenarios to predict responses accurately.
CT 9

1D, 2D, 3D (Higher dimension FEM)

1. Displacement in Higher Dimensions:

o In moving from 1D to 2D, displacement transitions from a scalar function to a vector,

incorporating components in both the xxx and yyy directions.

2. Strain Representation:

o Strain generalizes from a scalar in 1D to a second-order tensor in 2D and 3D,

accounting for both normal and shear strains in multiple planes.

3. Stress Tensor Complexity:

o The stress tensor expands from a scalar in 1D to a second-order tensor in 2D and 3D,
which includes normal and shear stresses acting on different planes.

4. Jacobian Matrix:

o The Jacobian generalizes from a scalar in 1D to a matrix involving partial derivatives

in higher dimensions, playing a crucial role in transforming coordinates and
integrating over the element.

5. Shape Functions:

o Shape functions are fundamental for interpolating values (like displacement and
geometry) between nodes within an element, facilitating the relationship between
local and global coordinates.

6. Dimensionality Effects:

o As dimensionality increases, additional strain and stress components appear,

requiring a more complex mathematical formulation to accurately describe the
behavior of materials under load.

7. Vector and Tensor Forms:

o The extension of FEM concepts to vector and tensor forms in higher dimensions is
necessary to capture the complexities of multi-dimensional behavior, which scalar
quantities cannot adequately represent.

8. Integration in FEM:

o The Jacobian matrix is critical for simplifying the integration of the stiffness matrix
and transforming the equations governing the behavior of elements.

9. Boundary Conditions:

o Understanding how to impose boundary conditions is essential, though the Jacobian

itself primarily affects the geometry and transformation of coordinates rather than
boundary conditions directly.

Key Topics for Review:

 Definitions and applications of stress and strain tensors

 Understanding and calculating Jacobians in various dimensions

 The role of shape functions in FEM

 Differences in stress and strain representation from 1D to 3D

 Practical applications of FEM in engineering problems

Transition from 1D to 2D

1. Coordinate System:

o 1D: In one dimension, we typically use a single coordinate xxx to represent the
position along a line (e.g., a beam).

o 2D: In two dimensions, we use two coordinates (x,y) to represent points in a plane
(e.g., a rectangular plate).

2. Displacement:

o 1D: Displacement is represented as a scalar function u(x).

o 2D: Displacement becomes a vector function ,

incorporating displacements in both the x and y directions.

3. Strain:

o 1D: Strain is a scalar, defined as

o 2D: Strain is represented as a 2D tensor:

o This includes normal strains ϵxx and ϵyy and shear strains ϵxy.

4. Stress:

o 1D: Stress is a scalar σ

o 2D: Stress becomes a 2D tensor:

o It incorporates normal stresses σxx and σyy and shear stresses σxy.

5. Jacobian Matrix:
 o The Jacobian matrix relates local coordinates to global coordinates and expands to
account for changes in geometry in 2D.

6. Shape Functions:

o In 2D, shape functions are used to interpolate the values of field variables (like
displacement) across an element, which is defined in local coordinates.

Transition from 1D to 3D

1. Coordinate System:

o 1D: A single coordinate xxx is used.

o 3D: Three coordinates (x,y,z) are used to represent points in space (e.g., a solid
cube).

2. Displacement:

o 1D: Displacement is a scalar function u(x)

o 3D: Displacement becomes a vector function

encompassing movements in the x, y, and z directions.

3. Strain:

o 1D: Strain is a scalar ϵ.

o 3D: Strain is represented as a 3D tensor:

o This includes normal strains along the three axes and shear strains.

4. Stress:

o 1D: Stress is a scalar σ.

o 3D: Stress becomes a 3D tensor:

o It incorporates normal stresses in all three directions and shear stresses acting on
different planes.

5. Jacobian Matrix:

o The Jacobian matrix becomes a 3x3 matrix that accounts for the transformation
from local coordinates to global coordinates in 3D.
6. Shape Functions:

o In 3D, shape functions are more complex and are used to interpolate values across
the volume of an element.

Summary

 The transition from 1D to 2D and 3D involves an increase in the number of coordinates and
degrees of freedom, leading to more complex representations of displacement, strain, and
stress.

 The use of tensors in higher dimensions allows for a more comprehensive analysis of
materials and structures subjected to various loading conditions.

 Understanding these transitions is crucial for accurately modelling and analysing mechanical
systems in engineering applications.