Practical Considerations

in Modelling
Dr. Aamir Mubashar
 Learning Objectives

• Discuss concepts that should be considered when

modeling
• Aspect ratio
• Symmetry
• Natural subdivisions
• Sizing of elements
• Methods of refinement
• Concentrated loads and infinite stress
• Infinite medium
• Connecting different kinds of elements
• To describe some of the approximations inherent in
finite element solutions
• To illustrate convergence of solution and introduce the
patch test for convergence of solution
 Learning Objectives

• To discuss the interpretation of stresses in an

element
• To present a flowchart of typical finite element
processes used for the analysis of plane stress and
plane strain
• To demonstrate various real-world applications
where plane stress/strain element models are
applicable
 Finite Element Modeling:
General Considerations

• Finite element modeling is

• partly an art
• guided by visualizing physical interactions taking place
within the body

• One appears to acquire good modeling techniques

through
• experience and
• by working with experienced people
Aspect Ratio and Element Shapes
• Aspect Ratio
• ratio of the longest dimension to the shortest dimension
of a quadrilateral element
Aspect Ratio and Element Shapes

• In many cases, as
the aspect ratio
increases, the
inaccuracy of the
solution increases
• Note that there
are exceptions to
this generalization
Aspect Ratio and Element Shapes

• General rules
• Try to maintain low aspect ratios
• Use corner angles of quadrilaterals near 90° when choosing
elements for a model
• Below are elements with shapes that should generally
be avoided
 Use of Symmetry / Antisymmetry

Symmetry line Antisymmetry line

Loads Loads

Symmetry requires correspondence in size, shape, and position of loads; material

properties; and boundary conditions that are on opposite sides of a dividing line or plane
 Use of Symmetry

Planar / Mirror Circular symmetry Repetitive symmetry

symmetry
 Use of Symmetry

X
 Use of Antisymmetry

X
 Use of Symmetry

• Mirror and reflective symmetry applications

 Use of Symmetry

• Mirror and reflective symmetry applications

Natural Subdivisions at Discontinuities

• Examples of using finite element discretization at

natural subdivisions
Natural Subdivisions at Discontinuities
 Mesh Revision (Refinement)

• Used to increase the accuracy of the results produced

by a finite element model
• First, start with a basic mesh using the fewest,
reasonable number of elements, and obtain a
benchmark result
• Then refine the model by doing actions like increasing
the mesh density and compare the results
• This process continues until the results converge to
some user-defined criteria
• Consider this original mesh for the
discussion of types of methods:
 h Method of Refinement

• In the h method, we use the particular element based

on the shape functions for that element
• Elements of the same kind are then added or made
smaller
• Uniform or non-uniform refinements are possible as
shown below:
 p Method of Refinement

• In the p method, the polynomial p is increased to

higher-order polynomials based on the degree of
accuracy specified
• These iterations are repeated until a convergence is
reached
 r Method of Refinement

• In the r method, the nodes are rearranged or

relocated without changing the number of
elements of the polynomial degree of the field
quantities
 Transition Triangles

• Triangular elements are useful for transitions from

smaller quadrilaterals to larger quadrilaterals
 Concentrated or Point Loads and
Infinite Stress

• Concentrated or point loads are usually expressed for

distributed loads as a mathematical convenience
• In the infinite element model, these loads would lead
to infinite stresses causing the material to yield
• Models can account for these concentrated loads by
highly refining the mesh or adding elements to account
for the deformation
 Infinite Medium

• The guideline for the finite element model is that

enough material must be included such that the
displacements at nodes and stresses within the
elements become negligible at locations far from
the foundation load

• The boundary
conditions must be
idealized to fit the
assumption of
infinite medium
Connecting (Mixing) Different Kinds of
Elements

• When connecting different kinds of elements, the

degrees of freedom of different elements must be
accounted for
• The figure below shows this for a beam and plane
element connected
• The calculated stresses in the plane element near
point A will normally be inaccurate
 Checking the Model for Errors

• Discretized model should be checked carefully for

errors by an objective party

• Many preprocessors detect obvious errors such as

• misplaced nodes
• missing elements
• misplaced loads
• misplaced boundary supports
 Checking Results

• Results should be checked for consistency by

making sure intended support nodes have zero
displacement
• Mesh convergence should be confirmed
• If symmetry exists then the results should exhibit
this symmetry
• FEM results should be compared with results from
other available techniques such as
• approximate material formulas
• experimental data
• and numerical analysis of simpler but similar problems
Equilibrium and Compatibility of Finite
Element Results

• Approximations generally inherent in finite element

solutions:
• Equilibrium of nodal forces and moments is satisfied
• Equilibrium within an element is not always satisfied
• Equilibrium is not usually satisfied across interelement
boundaries
• Continuity is satisfied within an element as long as the
element displacement field is continuous
• In the formulation of the element equations, continuity
is invoked at the nodes
• Continuity may or may not be satisfied along
interelement boundaries
 Convergence of Solution and Mesh
Refinement

• Compatible displacement functions are necessary

in producing a finite element model which
converges to a solution
• There is a lower bound on the displacement of the
problem using complete displacement functions as
seen below:
 Patch Test

• The Patch test is used to test the convergence of a

solution of an element being used in the model
• The test requires that the element must be able to
accommodate both rigid-body motion and a
constant state of strain, as both are possible within
a structure
• The test can also be applied to determine if
sufficient Gauss points have been used in the
numerical integration process to evaluate the
stiffness matrix for isoparametric formulations
 Patch Test

• Considering a simple finite

element model composed of four
irregular shaped elements of the
same material with at least one
node inside of the patch
• The elements should be irregular, Displacement Patch Test
as some regular elements (such as
rectangular) may pass the test
• Perform “displacement” patch test
• Perform “Constant Strain” patch
test
Force Patch Test
 Interpretation of Stresses

• The best approximation of the stress occurs at the

midpoint of elements as shown in the figure
• For higher-order elements like the linear-strain
triangle, the common practice is to directly
evaluate the stresses at the centroid of the element
• An alternative
procedure is to use an
average value of the
stresses evaluated at
each node. This is
called smoothing
 Summary

• Discussed general considerations in finite element modeling

• Discussed the effects of aspect ratios and element shape
• Illustrated the use of symmetry
• Showed examples of natural subdivisions and discontinuities
• Reviewed the h, p, and r methods of mesh refinement
• Discussed the use of transition triangles
• Discussed how to treat concentrated or point loads
• Reviewed how to represent an infinite medium
• Discussed how to connect different kinds of elements
• Reviewed how to check a model for errors and checking the results
• Discussed equilibrium compatibility and interpretation of stresses
• Showed a flowchart of plane stress/strain finite element process
THANK YOU