My Common Notes on

Finite Element
Analysis
 FEA CFD
Feature FEA (Finite Element Analysis) CFD (Computational Fluid Dynamics)

A simulation method used to analyze A simulation method used to analyze

Definition stress, deformation, and structural fluid flow, heat transfer, and
behavior of solid objects. aerodynamics in liquids and gases.

How a solid structure reacts to forces, How fluids (air, water, gases) move and
What It Studies
loads, heat, and vibrations. interact with surfaces.

Solves structural mechanics Solves fluid flow equations like Navier-

Main Equations Used equations like Newton’s laws and Stokes equations and continuity
Hooke’s law. equations.

Material Type Solids (metals, plastics, composites). Fluids (liquids, gases, air).

Stress, strain, deformation, fatigue, Pressure distribution, velocity,

Primary Outputs vibration, and thermal expansion temperature, turbulence, and flow
results. patterns.

Used for automobile chassis, aircraft Used for aerodynamics, cooling

wings, machine components, and systems, combustion engines, HVAC,
Typical Applications
building structures to check and medical devices to optimize fluid
durability and safety. behavior.

Tetrahedral, hexahedral, polyhedral,

Tetrahedral, hexahedral, shell, beam
Mesh Type Used and structured/unstructured grids (fine
elements (structured or unstructured).
mesh needed for turbulence).

Applied as inlet/outlet flow, pressure,

Applied as forces, constraints,
Boundary Conditions temperature, turbulence conditions on
temperatures on solid structures.
fluid regions.

Higher, as fluid motion is highly

Computational Lower than CFD, as solid structures
dynamic and requires solving many
Complexity do not change shape dramatically.
variables over time.

ANSYS Mechanical, Abaqus, ANSYS Fluent, OpenFOAM, SimScale,

Software Examples
SolidWorks Simulation, NASTRAN. COMSOL Multiphysics.

Used in automotive, aerospace, Used in aerospace, automotive, HVAC,

Industry Usage construction, robotics, and biomedical, energy, and chemical
biomechanics for strength analysis. industries for fluid and heat analysis.

Structural stability, mechanical

Fluid flow, heat transfer, aerodynamics,
Best For strength, and thermal expansion
and combustion simulations.
analysis.
 Image: Monti, Stefano. (2018). Reduction
of the Local Stress Field around Holes
through Porous Shaped Structures.
Communications - Scientific Letters of the
University of Zilina. 20. 36-41.
10.26552/com.C.2018.3.36-41.

Von Mises Principal Shear

Stress Stress Stress
Feature Von Mises Stress Principal Stress Shear Stress

Equivalent stress used Maximum and minimum Stress acting

Definition to predict yielding in normal stresses acting tangentially to a
ductile materials on a plane surface or plane

Scalar value derived Normal stress (acts Tangential stress

Type of Stress from combined stress perpendicular to a (acts parallel to a
components surface) surface)

Pascals (Pa), MPa, or Pascals (Pa), MPa, or Pascals (Pa), MPa, or

Units
N/mm² N/mm² N/mm²

Yield criteria (Von Mises Stress analysis and Failure analysis in

Used In
yield theory) failure prediction torsion, bolts, shafts

Orientation and
If a ductile material will Points of maximum
Helps Determine magnitude of max/min
yield or not sliding or distortion
stresses

Torsional stress in
Application Pressure vessel, Pressure vessels,
shafts, pins, and
Example crankshaft, brackets structural beams
rivets

Yes, widely used for Yes, especially in

Yes, used for evaluating
Common in FEA? post-processing and dynamic and
stress extremes
safety checks torsional cases

τ plots or vectors
Graphical Scalar contour plot Individual plots for σ₁, σ₂,
indicating shear
Output in FEA (color-coded field) σ₃ (principal directions)
planes
 Implicit FEA Explicit FEA

Feature Implicit FEA Explicit FEA

A step-by-step calculation A highly detailed simulation

method used for steady, slow, or method used for fast, highly
Definition linear problems like static dynamic, or nonlinear problems
analysis. like crash testing.

Larger time steps, making it Very small time steps, required for
Time Step Size suitable for problems with fast-changing and impact-based
gradual changes. problems.

Faster for short-duration, high-

Slower for dynamic problems, but
Computation Speed faster for steady-state analysis.
speed events, but
computationally intensive.

Used for static and low-speed

Used for high-speed impact,
dynamic problems (e.g., stress
Problem Type analysis in a bridge, thermal
explosions, and crash simulations
(e.g., car crash analysis).
expansion).

Uses matrix-based iterative Solves one step at a time, making

Solution Method solving to balance forces over it ideal for problems with extreme
time. deformations.

Requires less memory and Requires high computational

Memory & Processing processing power, making it power and memory, as it
efficient for large models. calculates every tiny step in detail.

Bridge load testing, pressure

Car crash testing, metal forming,
Examples vessel stress analysis, bolt
drop tests, explosion simulations.
tightening simulation.

Slow or continuous force Fast, impact-driven, or highly

Best For applications with stable nonlinear events where materials
conditions. behave unpredictably.
 Image: Kim, Ki-Tae & Zhang, Lingbo & Bathe, Klaus-Jürgen. (2018). Transient
implicit wave propagation dynamics with overlapping finite elements. Computers
and Structures. 199. 18-33. 10.1016/j.compstruc.2018.01.007.

Structured Mesh Unstructured Mesh

Feature Structured Mesh Unstructured Mesh

Regular, grid-like pattern (rows and Irregular, free-form arrangement of

Mesh Arrangement
columns) elements

Typically quadrilaterals (2D) or Typically triangles (2D) or

Element Shape
hexahedrals (3D) tetrahedrals (3D)

Requires additional data to define

Mesh Connectivity Easily defined using indices
connectivity

Ideal for complex and irregular

Geometry Handling Best for simple, regular geometries
geometries

Generally higher with fewer elements May need more elements to achieve
Accuracy
(if well aligned) similar accuracy

Faster and more automated mesh

Mesh Generation Time Time-consuming for complex shapes
generation

More efficient and faster due to Less efficient, may require more
Solver Efficiency
regular structure memory and time

Allows local refinement easily

Refinement Control Local refinement is more difficult
(adaptive meshing)

Harder to visualize due to irregular

Visualization & Debugging Easier to visualize and debug
patterns

CFD in ducts, rectangular domains, Complex geometries like automotive

Common Applications
simple FEA models parts, biomedical models
 Image: https://www.comsol.com/
Mesh Refinement Adaptive Meshing Coarse Meshing

Feature Mesh Refinement Adaptive Meshing Coarse Meshing

The process of manually

Automatically adjusts the Uses a lower number of
increasing the number of
mesh density based on elements to reduce
Definition elements in specific
error estimation during computational cost but
regions to improve
simulation may sacrifice accuracy
accuracy

Varies dynamically,
High in selected areas, Low density, fewer
Mesh Density adjusted during the
manually refined elements and nodes
simulation

Optimized, balances Lower accuracy, suitable

High, especially in critical
Accuracy accuracy and for rough
stress or flow regions
computational cost approximations

Increases as more Moderate, as the mesh

Computation Faster, since fewer
elements require more changes dynamically for
Time elements are calculated
processing efficiency

Simulations where
Detailed studies where Quick preliminary
automated optimization
high precision is needed, analysis where
Best For can improve accuracy,
such as stress approximate results are
such as fluid flow or
concentration areas sufficient
dynamic problems

Manual, requires
Automatic, controlled by Manual, but limited to
User Control engineering judgment to
software algorithms defining a basic mesh
refine areas

Structural analysis, CFD CFD simulations, Concept testing, large-

Common
near walls, contact stress nonlinear FEA, transient scale simulations with
Applications zones problems limited resources
 FDM FEM FVM
Finite Difference Method Finite Element Method Finite Volume Method
Aspect
(FDM) (FEM) (FVM)

Uses a variational (or weighted Integrates the governing

Approximates derivatives by
residual) approach to transform equations over control volumes,
Fundamental Concept using differences at discrete grid
the governing equations into a ensuring the conservation of
points.
weak form. fluxes locally and globally.

Employs unstructured (or Divides the domain into control

Typically relies on structured
structured) meshes that can volumes which can be arranged
grids (regularly spaced points),
Domain Discretization which limits its flexibility for
adapt to complex geometries, in structured or unstructured
using elements like triangles, meshes, offering flexibility in
complex geometries.
quadrilaterals, tetrahedrons, etc. handling complex domains.

Converts the strong form of the Applies the integral form of the
Directly approximates differential
differential equations into a weak governing equations, ensuring
equations using finite
Governing Equations differences, converting them into
form, making it suitable for that conservation laws (mass,
complex boundary conditions momentum, energy) are satisfied
algebraic equations.
and irregular domains. within each control volume.

Conservation is achieved Naturally conserves fluxes across

Not inherently conservative; may
indirectly through the variational control volume faces, making it
Conservation require additional treatment to
formulation, but care is needed highly suitable for fluid dynamics
ensure flux conservation.
to maintain local conservation. and related conservation laws.

Well-suited for complex

Highly flexible and capable of
Best suited for simple, regular geometries; unstructured control
Handling Complex accurately modeling complex,
geometries due to its reliance on volumes allow accurate
Geometries irregular geometries through
structured grids. representation of curved or
adaptive meshing.
intricate boundaries.

Moderate complexity; requires

More complex due to the need
integration over control volumes
Relatively simple to implement on for mesh generation, element
Implementation and careful treatment of fluxes,
uniform grids; straightforward formulation, and assembling the
Complexity but benefits from inherent
finite difference approximations. system of equations using shape
conservation properties which
functions.
are crucial in CFD.

Structural analysis, solid Computational Fluid Dynamics

Heat transfer problems, simple
mechanics, complex PDEs, and (CFD), multiphase flows, and any
fluid flows, and problems where
Typical Applications high-order accuracy on
problems where geometric applications where conservation
flexibility and precision are of mass, momentum, and energy
structured domains is sufficient.
required. are critical.
 Image: Cheng, Ding-Shing & Hung, Cheng-Wei & Pi,
S.-J. (2013). Numerical simulation of near-field
explosion. Journal of Applied Science and
Engineering. 16. 61-67.
Lagrangian Approach Eulerian Approach ALE

Arbitrary Lagrangian-
Feature Lagrangian Approach Eulerian Approach
Eulerian (ALE)

Mesh motion is partially

Mesh moves with the Mesh is fixed; material
Basic Concept independent of material
material flows through it
flow

Mesh deforms with the Mesh can move or deform

Mesh Behavior Mesh remains stationary
material arbitrarily

Fluid flow, especially with Fluid-structure

Solid mechanics and
Best For large deformation or free interaction, large
structural analysis
surface deformation problems

Tire simulations, piston-

Crash simulations, metal CFD simulations, sloshing,
Common Applications cylinder systems, FSI
forming, structural FEA combustion
problems

Combines benefits of both

Tracks material No mesh distortion due to
Advantages Lagrangian and Eulerian
boundaries accurately fixed mesh
methods

Mesh distortion under Difficult to track material More complex to

Limitations
large deformation interfaces implement and solve

Easy (since mesh follows Difficult (requires Moderate (better control

Material Tracking
material) additional techniques) over tracking)

Computational Efficient for small Efficient for fluids and More computationally
Efficiency deformations large flow domains intensive

ANSYS, Abaqus (ALE

Fluent, OpenFOAM, STAR-
Used In Software Abaqus Explicit, LS-DYNA modules), COMSOL
CCM+
Multiphysics
 Image: https://www.idealsimulations.com/resources/
turbulence-models-in-cfd/
Reynolds-Averaged Large Eddy Direct Numerical
Navier-Stokes Simulation Simulation

RANS (Reynolds-Averaged DNS (Direct Numerical

Feature LES (Large Eddy Simulation)
Navier-Stokes) Simulation)

Solves time-averaged Resolves large eddies,

Resolves all turbulence
Approach flow equations; models all models only small-scale
scales without modeling
turbulence turbulence

Moderate (depends on High (captures transient Very high (captures full flow
Accuracy
turbulence model) large-scale structures) physics)

High (needs fine mesh and Very high (requires extreme

Computational Cost Low (most efficient)
small time steps) computational resources)

Fundamental turbulence
Industrial flows, HVAC, Complex flows like
Use Case research, small-scale
automotive, etc. combustion, vortex shedding
studies

Steady or unsteady Unsteady (resolves flow Unsteady (fully time-

Time Dependence
depending on setup structures over time) resolved simulation)

Extremely fine mesh (down

Mesh Requirements Coarse to moderate Fine mesh for large eddies
to Kolmogorov scales)

Turbulence Empirical models (e.g., k-ε, No turbulence modeling

Sub-grid scale (SGS) models
Modeling k-ω, SST) required

Typical Software Fluent, OpenFOAM, Fluent, OpenFOAM, STAR- Research codes, some DNS-
Support ANSYS CFX, STAR-CCM+ CCM+, SimScale specific solvers

Flow over vehicles, ducts, Jet flows, combustion, Detailed study of turbulent
Example Application
pipes external aerodynamics boundary layers
 Linear Nonlinear Dynamic
Analysis Analysis Analysis

Feature Linear Analysis Nonlinear Analysis Dynamic Analysis

Assumes proportional Considers large deformations,

Studies time-dependent
Definition relationship between loads material nonlinearities, and
behavior under loads
and deformations contact effects

Changes due to material Can be linear or nonlinear,

Stress-Strain Remains constant (Hooke’s
yielding or geometric depending on the loading
Relationship Law applies)
deformation conditions

Small loads where Large loads causing plasticity Can involve static or time-
Load Magnitude
deformation is minimal or instability varying loads

Elastic (does not change with Can be elastic, plastic, or Can involve vibration,
Material Behavior
load) hyperelastic impact, or harmonic motion

Accounts for large

Geometric deformations and instability Considers mass, stiffness,
Assumes small deformations
Considerations effects (buckling, snap- and damping effects
through, etc.)

Boundary Remain constant throughout Can change during the Can vary with time and
Conditions the analysis simulation due to deformation frequency

Time-independent (static Time-independent (but Time-dependent (dynamic

Time Dependency
analysis) accounts for nonlinear effects) response is crucial)

Computational Lower, since equations Higher, due to iterative Higher, as it involves time or
Cost remain simple solution methods frequency domain analysis

Modal analysis, Time-history

Common Methods Direct stiffness method, Newton-Raphson method,
analysis, Harmonic
Used Superposition principle Incremental-iterative solvers
response analysis

Vibrations in bridges,
Small deflections in beams, Large deformations in rubber
Applications pressure vessel analysis seals, buckling in thin plates
seismic analysis of
buildings, crash simulations
 Image: Haertel, Jan. (2017). Design of Thermal
Systems Using Topology Optimization.

Topology Shape Size

Optimization Optimization Optimization

Feature Topology Optimization Shape Optimization Size Optimization

Optimize the material Modify the external Adjust dimensions like

Objective layout within a design boundaries of a thickness, length, or
space component cross-section

Outer contour and Parameter values like

Internal structure and
What Changes? surface shape of the beam thickness or bolt
material distribution
model diameter

Very high (starts with full Moderate (fixed topology, Limited (no change in
Design Freedom
material volume) adjustable boundaries) shape or topology)

High (requires multiple

Computational Effort Moderate Low
iterations and fine mesh)

Aerodynamic shaping,
Lightweight structures, Structural sizing, cost
Common Applications stress concentration
aerospace, automotive and weight reduction
reduction

Requires post-processing Easier to implement in Directly modifiable in

CAD Integration
or remodeling for CAD CAD parametric CAD models

Organic, lattice-like Smooth boundary Same geometry with

Result Type
structures changes optimized dimensions

Design refinement based

Material-efficient design, Performance fine-tuning
Best For on performance or
conceptual stages of geometry
constraints
 Material Geometric Contact
Nonlinearity Nonlinearity Nonlinearity

Feature Material Nonlinearity Geometric Nonlinearity Contact Nonlinearity

Occurs when
Occurs when material Arises when two or more
deformations are large
behavior does not follow bodies come into or out
Definition enough to affect
Hooke’s Law (stress- of contact during
geometry during
strain curve is nonlinear) deformation
analysis

Plasticity, creep, Contact between

Large deformations,
Cause hyperelasticity, surfaces with friction,
large rotations, buckling
viscoelasticity separation, or sliding

Requires nonlinear Requires contact

Requires nonlinear
material models (e.g., definitions and
Modeling Requirement strain-displacement
plastic, rubber, algorithms to handle
relationships
viscoelastic) interaction

Metal forming, rubber Soft structures, buckling Gear analysis, bolted

Common in components, creep of thin walls, crash joints, assemblies,
analysis simulation impact problems

May still be linear or Depends on the nature

Stress-Strain
Nonlinear (curve) nonlinear depending on of contact (e.g., frictional
Relationship
material or frictionless)

High (due to
Analysis Complexity Moderate to High Moderate to High
convergence issues)

Needs material models Enable geometric Define contact pairs,

Software Handling like plasticity, nonlinearity in solver types (bonded, frictional,
hyperelastic, etc. (large deformation flag) etc.), and solver settings

Bending of a flexible
Yielding of steel beyond Simulation of a bolted
Example beam or balloon
elastic limit connection under load
inflation