Tribhuvan University

Institute of Engineering (IOE)

Finite Element Method

Lecturer: Kamal Darlami

darlami.kd@pcampus.edu.np
 Introduction to the Finite Element course
Lecture objectives
 Understand course objectives
 References and assessment
 History of FEM
 Finite Element Capabilities
 Overview of the FE modelling process

Module objectives
 On completion of this module you should:
understand the basis of the Finite Element Method
understand the basis of finite element programming (with MATLAB)
be able to derive finite element methods for the numerical solution of simple differential equations

 You should also be able to set up and solve various simple finite element models manually
1D axially loaded bars, 1D thermal flow through a wall, 2D steady state heat flow, Torsion of
prismatic bars, 2D irrotational fluid flow, 2D plane stress and plane strain analysis and
axisymmetric solid mechanics problems
 Introduction to the Finite Element course
Reference and Reading Material (in google drive)
 Course Notes
 Exercise sheet
 Computer codes
 Course text
•TP Chandrupatla & AD Belegundu, Introduction to finite elements in engineering, Prentice-Hall, Englewood Cliffs, N.J., 1991.

Recommended reading
• Zienkiewicz, O.C.; Morgan, K., Finite Elements and Approximations, Wiley, 1983
• J Fish & T Belytschko. A first course in finite elements, Wiley, Chichester, 2007.
• RD Cook. Finite element modelling for stress analysis, Wiley, Chichester, 1995.
• DS Burnett. Finite element analysis: from concepts to applications, Addison-Wesley, Reading, Massachusetts, 1987.
• E Hinton & DRJ Owen, Finite Element Programming, Pineridge Press, Swansea, 1977.

Check google drive regularly

•Updates to the course notes will be uploaded if needed
•The assignment will be published in due course
 Introduction to the Finite Element course
Assessment
Group project –Matlab Programming (20%)
Deadline: Last teaching week
Internal examination (10 %)
Mid-semester
Laboratory Work/Report (10%)
Using Commercial Software
Final examination (60%)
End of semester
Closed book
3 hours duration
Overall pass mark 50%
 Finite Element Method Defined
• Complexities in the geometry, properties and in the boundary conditions that
are seen in most real-world problems usually means that an exact solution
cannot be obtained or obtained in a reasonable amount of time.
• Engineers are content to obtain approximate solutions that can be readily
obtained in a reasonable time frame, and with reasonable effort. The FEM is
one such approximate solution technique.
• The FEM is a numerical procedure for obtaining approximate solutions to
many of the problems encountered in engineering analysis.
• The continuum has an infinite number of degrees-of-freedom (DOF), while the
discretized model has a finite number of DOF. This is the origin of the name,
finite element method.
• The number of equations is usually rather large for most real world
applications of the FEM, and requires the computational power of the digital
computer. The FEM has little practical value if the digital computer were not
available.
• Solution of FEM gives the approximate behavior of the continuum or system.
 History of FEM
 Hernikoff (1941): used a lattice of line elements for solving stress in continuous solids
 Courant (1943): used the triangular elements and energy method to study the torsion problem
 Levy(1947): developed the flexibility or force method in structural analysis
 Levy (1953): presented another method called the stiffness or displacement method in analyzing aircraft structures
 Argyris & Kelsey (1954): developed matrix structural analysis method using energy principles
 Turner, Clough, Martin, Topp (1956): derived stiffness matrices for truss, beam and 2-D elements
 Clough (1960): first used the term “finite element”
 Melosh (1961): developed a flat rectangular-plate bending element
 Grafton & Strome (1963): developed the curved-shell bending element
 Gallagher & Padlog (1963): initially treated the buckling problems
 Archer (1965): considered dynamic analysis in bar and beam structures
 Zienkiewicz &Cheung (1965), Martin (1968), Wilson and Nickel (1966): studied torsion of shaft, fluid flow, and heat
conduction
It is difficult to document the exact origin of the FEM, because the basic concepts have evolved over a period of 150
or more years. The first book on the FEM by Zienkiewicz and Chung was published in 1967.
 History of FEM
Engineering Mathematics
Finite Difference
Trail Function
Method
Weighted Residual
Variational Method
Method
Rayleigh 1870 Gauss 1795
Similar Structure Ritz 1909 Galerkin 1915
Biezeno-Koch 1923
Replacement
Continuous Trail
Hrenikoff 1941
Function Variable Finite
Machenry 1943
Newmark 1949 Courant 1943 Difference Method
Prager-Synge 1947 Varga 1962
Direct Continuum
Zienkiewicz 1964
Elements
Argyris 1955
Turner et al. 1956
Present Finite
Element Method
First coined by Clough
1960
Commercially available general FEM software

FEM
softwares
 Finite Element Analysis Capabilities
Linear static Stress Analysis
Factor of Safety Calculation
Part & Assembly Stress Analysis
Deflection Calculations
Correlation to Measurements of Deflections and Strains
Contact Stress Computation
Super-position of Thermal Stresses
Stiffness Calculations to achieve stated Targets

Frequency & Buckling Analysis

Computation of Frequencies & Mode Shapes
Modal Assurance Criteria (MAC)
Correlation to Measured data
Buckling Calculations for axially loaded members
Critical Speed Calculations
Campbell Diagram for Rotor-dynamics
Point Mobility Analysis
 Finite Element Analysis Capabilities
Analysis of Composites
Failure mode prediction of Composite panels
Filament Wound Composite – Anisotropic material modeling
Random Fiber Composites
Stiffness, Deflection and Critical Load calculation of Composite Structures
Metal Matrix Composites – Thermo mechanical Analyses

Dynamic Analysis
Frequency Response Analysis
Seismic Analysis Response Calculations
Harmonic Analysis
Random Vibration Calculations
Dynamic Stress Computations
Power Train Vibration Analysis
Shock Calculations per NAVSEA, DDAM, MIL STD
 Finite Element Analysis Capabilities
Non-Linear Analysis
Material Non-linear Analysis
Geometric Non-linear Analysis
FEA of Rubber & Elastomers
Non-linear Dynamic Analysis
Time Domain Response Analysis
Impact Analysis
Thermo-mechanical Analysis involving large displacements
Elasto-plastic Deformation Analysis

Fatigue Analysis
Remaining Life Analysis ( RLA )
Durability Analysis
Failure Prediction Analysis
High Cycle Fatigue Calculations
Correlation to Real-world situations
Comparison of Alternate materials for extended life and warranty
Life extension analysis
 Finite Element Analysis Capabilities
ASME Stress Analysis
Stress Analysis per ASME Codes
Nozzle stress analysis
Stress Intensity Calculations
Shell & Full Scale 3D Stress Analysis of Pressure Vessels among others

Design Optimization
Optimization of CAD Geometries
Weight Reduction Analysis
Value Addition & Value Engineering Analysis
Sensitivity Based Optimization
Optimization of design variables based on performance targets
 Finite Element Analysis Capabilities
Thermal Analysis
Thermal Stress Analysis of parts and assemblies
Transient Thermal Analysis
Thermo-mechanical Analysis
Coupled Thermo-fluid analysis
Natural and Forced Convection Analysis
Non-Linear Thermal analysis of curing processes
Creep Analysis

CFD Fluid Flow Analysis

Pressure Drop Calculations
Conjugate Heat Transfer Analysis
Electronic Cooling Analysis
Thermal Efficiency Calculations
Fluid Flow simulation in Devices such as pumps, valves, ducts, piping networks, fans,
diffusers, cyclones, blowers, heat exchangers
Design optimization based on performance prediction
 FEM - powerful tool
How can the FEM Help the Design Engineer?
• The FEM offers many important advantages to the design engineer:
• Easily applied to complex, irregular-shaped objects composed of several different materials and having complex boundary conditions.
• Applicable to steady-state, time dependent and eigenvalue problems.
• Applicable to linear and nonlinear problems.
• One method can solve a wide variety of problems, including problems in solid mechanics, fluid mechanics, chemical reactions,
electromagnetics, biomechanics, heat transfer and acoustics, to name a few.

How can the FEM Help the Design Organization?

Simulation using the FEM also offers important business advantages to the design organization:
• Reduced testing and redesign costs thereby shortening the product development time.
• Identify issues in designs before tooling is committed.
• Refine components before dependencies to other components prohibit changes.
• Optimize performance before prototyping.
• Discover design problems before litigation.
• Allow more time for designers to use engineering judgment, and less time “turning the crank.”
 Advantages of the Finite Element Method
 Can readily handle complex geometry:  Can handle bodies comprised of nonhomogeneous materials:
• The heart and power of the FEM. • Every element in the model could be assigned a different
 Can handle complex analysis types: set of material properties.
• Vibration  Can handle bodies comprised of nonisotropic materials:
• Transients • Orthotropic
• Nonlinear • Anisotropic
• Heat transfer  Special material effects are handled:
• Fluids • Temperature dependent properties.
 Can handle complex loading: • Plasticity
• Node-based loading (point loads). • Creep
• Element-based loading (pressure, thermal, • Swelling
inertial forces).  Special geometric effects can be modeled:
• Time or frequency dependent loading. • Large displacements.
 Can handle complex restraints: • Large rotations.
• Indeterminate structures can be analyzed. • Contact (gap) condition.
 Disadvantages of the Finite Element Method
 A specific numerical result is obtained for a specific problem. A general closed-form solution, which would permit one to examine
system response to changes in various parameters, is not produced.
 The FEM is applied to an approximation of the mathematical model of a system (the source of so-called inherited errors.)
 Experience and judgment are needed in order to construct a good finite element model.
 A powerful computer and reliable FEM software are essential.
 Input and output data may be large and tedious to prepare and interpret.
 Numerical problems:
• Computers only carry a finite number of significant digits.
• Round off and error accumulation.
• Can help the situation by not attaching stiff (small) elements to flexible (large) elements.
 Susceptible to user-introduced modeling errors:
• Poor choice of element types.
• Distorted elements.
• Geometry not adequately modeled.
 Certain effects not automatically included:
• Buckling
• Large deflections and rotations.
• Material nonlinearities .
• Other nonlinearities.
 Future Trends in the FEM and Simulation
 The FEM in particular, and simulation in general, are becoming integrated with the entire product development process (rather than just
another task in the product development process).
 A broader range of people are using the FEM.
 Increased data sharing between analysis data sources (CAD, testing, FEM software, ERM software.)
 FEM software is becoming easier to use:
 Improved GUIs, auto meshers.
 Increased use of sophisticated shell scripts and wizards.
 Enhanced multi-physics capabilities are coming:
 Coupling between numerous physical phenomena.
» Ex: Fluid-structural interaction is the most common example.
 Increasing use of non-deterministic analysis and design methods:
 Statistical modeling of material properties, tolerances, and anticipated loads.
 Sensitivity analyses.
 Faster and more powerful computer hardware. Massively parallel processing.
» Ex: ADVENTURE PROJECT @ the University of Tokyo.
 Decreasing reliance on testing.
 FEM and simulation software available freely.
» Ex: OpenSees @ University of California, Berkeley .
» Ex: ADVENTURE PROJECT @ the University of Tokyo.
 Introduction to the Finite Element course
A typical finite element analysis project

A problem must be Plan an initial FE

solved Anticipate physical model. Pre-process:
behavior. prepare the FE
model.
Plan how FE results
Yes Plan a revised FE
Is FE analysis will be checked to
model using insight
required? see if they are
provided by the Solve equation of
reasonable.
No current FE model. the FE model.

Analytical or
No
experimental
solution Post-process:
Are results reasonable? display FE results
Are error estimates small?
Yes Does model revision do little to alter computed results?
Stop Computer Software
 Introduction to the Finite Element course
Physical Problem Change of physical problem

Improve mathematical model

Mathematical model
Governed by differential equations
Assumptions on
• Geometry Finite element solution
• Kinematics Choice of
• Material law • Finite elements
• Loading • Mesh density
• Boundary Conditions • Solution parameters Refine mesh
Representation of solution
• Loading parameters
• Boundary condition Refine Design improvements
etc.
• Etc. analysis Structural optimization
Finite element
analysis process Assessment of accuracy of finite element
Interpretation of results
solution of mathematical model
Finite element solution of mathematical model
 Introduction to the Finite Element course
Example: Steel load bearing component 𝑀𝑚𝑎𝑥
Maximum stress 𝜎𝑚𝑎𝑥 = 𝑦𝑚𝑎𝑥
𝐸𝐼
Maximum deflection
2𝑃𝑎2 𝑏 3 2𝑏𝐿
𝑊𝑚𝑎𝑥 = 𝑎𝑡 𝑥 = 𝐿 −
Beam theory: The simplest model 3𝐸𝐼 3𝑏 + 𝑎 2 3𝑏 + 𝑎
(a < b is assumed)
Plane stress theory: Finite element approximation

𝑃𝑎𝑏 2 𝑃𝑎2 𝑏
𝐿2 𝐿2
Finite element mesh

𝑃𝑎2 𝑏 𝑃𝑎2 𝑏
𝑀𝑚𝑎𝑥 = 2 − 3 (3𝑎 + 𝑏)
𝐿 𝐿
Deflections Effective stress
 Introduction to the Finite Element course
Equilibrium partial differential equations
𝜕𝜎𝑥𝑥 𝜕𝜏𝑥𝑦 𝜕𝜏𝑦𝑥 𝜕𝜎𝑦𝑦
+ = 0, + =0
𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦
Stress-strain relation Strain-displacement relation
𝜎𝑥𝑥 1 𝜈 0 𝜕𝑢
𝐸
𝜎𝑦𝑦 = 𝜈 1 0 𝜀𝑥𝑥 =
𝜏𝑥𝑦
2
1−𝜈 0 0 𝜕𝑥
(1 − 𝜈)/2
𝜕𝑣
𝜀𝑦𝑦 =
Boundary conditions: 𝜎𝑛𝑛 = 0, 𝜏𝑛𝑡 = 0 𝜕𝑦

• Except at the surface of the holes where zero displacement is imposed 𝜕𝑢 𝜕𝑣

𝑢=𝑣=0 𝛾𝑥𝑦 = +
𝜕𝑦 𝜕𝑥
• And at a point of load where 𝜎𝑛𝑛 has to be imposed
 Trial solution procedures for differential equations
Lecture objectives
• Review the Trial Solution Procedure used to find approximate solutions to differential equations
• Required to understand the Finite Element Method
Example
• Consider the second-order differential equation

𝑑 𝑑𝑢
𝐸𝐴 +𝑓 𝑥 =0
𝑑𝑥 𝑑𝑥
Subject the boundary conditions

𝑑𝑢
𝑢 0 = 0; 𝐸𝐴 =𝑃
𝑑𝑥 𝑥=𝑙

The main goal is to find an approximate solution to this problem!

 Trial solution procedures for differential equations
General trial solution procedure
1. Construct a trial solution function dependent upon a set of scalar parameters
2. Define a suitable optimising criterion for the solution (weighted residual method)
3. Determine the parameters of the trial solution according to the optimising criterion
4. At the end, you may proceed to estimate the accuracy of the solution

Trail solution construction

For example, choose 𝑢 𝑥; 𝐚 = 𝜙0 𝑥 + 𝑎1 𝜙1 𝑥 + 𝑎2 𝜙2 𝑥 + … … … … + 𝑎𝑛 𝜙𝑛 (𝑥)
Trail functions: 𝜙0 , 𝜙1 , 𝜙2 , … … … … … … , 𝜙𝑛
DOF (Scalar parameters): 𝑎1 , 𝑎2 , 𝑎3 , … … … … … , 𝑎𝑛
 Trial solution procedures for differential equations
Optimising criteria
• We want to find the values for 𝑎1 , 𝑎2 , 𝑎3 , … … , 𝑎𝑛 that gives the best approximation to the differential equation
• There are different criteria to define the best approximation
• We will study the following methods of weighted residuals
# Point collocation # Subdomain collocation
# Least squares # Galerkin method
Methods of weighted residuals
𝑑 𝑑𝑢
• To define an optimising criterion, we define the residual function 𝑟 𝑥; 𝐚 = 𝑘 −𝑓
𝑑𝑥 𝑑𝑥

• Where 𝐚 = 𝑎1 , 𝑎2 , 𝑎3 , … … , 𝑎𝑛 and 𝑢(𝑥) = 𝜙0 𝑥 + 𝑎1 𝜙1 𝑥 + 𝑎2 𝜙2 𝑥 + … … … … + 𝑎𝑛 𝜙𝑛 (𝑥)

• If 𝑟 = 0, then the trial solution is the exact solution. But this is not possible in general
• We can find a set of coefficients that minimise the residual. This set will define an approximate (numerical) solution to the
problem.
• We will study different criteria (optimising criteria) to define the best approximation
 Trial solution procedures for differential equations
Methods of weighted residuals
Point Collocation Method Subdomain Collocation Method
• Choose a number of points (collocation points) • Choose a number of intervals (subdomains)
• Solve the system of equations • Solve the system of equations
Δ𝑥1 Δ𝑥2 Δ𝑥𝑛
x = a 𝑥1 𝑥2 𝑥𝑛 x = b x=a x=b
.
𝑟 𝑥1 ; 𝐚 = 0 𝑟 𝑥; 𝐚 𝑑𝑥 = 0
Δ𝑥1
𝑟 𝑥2 ; 𝐚 = 0 .
𝑟 𝑥; 𝐚 𝑑𝑥 = 0
𝑟 𝑥3 ; 𝐚 = 0 Δ𝑥2
| |
| |
.
𝑟 𝑥𝑛 ; 𝐚 = 0 𝑟 𝑥; 𝐚 𝑑𝑥 = 0
Δ𝑥𝑛
 Trial solution procedures for differential equations
Methods of weighted residuals
Least Squares Galerkin Method
.
• Find the coefficients so that Ω 𝑟 2 𝑥; 𝐚 𝑑𝑥 • Find the coefficients so that the weighted averages
is minimum of the residual over the entire domain vanish
• Solve the system of equations • We will see how the weighting functions are
generated by means of finite elements
. .
𝜕
𝑟 2 𝑥; 𝐚 𝑑𝑥 = 0 𝜙1 𝑥 𝑟 𝑥; 𝐚 𝑑𝑥 = 0
𝜕𝑎1 Ω Ω
. .
𝜕
𝑟 2 𝑥; 𝐚 𝑑𝑥 = 0 𝜙2 𝑥 𝑟 𝑥; 𝐚 𝑑𝑥 = 0
𝜕𝑎2 Ω Ω
| |
| |
. .
𝜕
𝑟 2 𝑥; 𝐚 𝑑𝑥 = 0 𝜙𝑛 𝑥 𝑟 𝑥; 𝐚 𝑑𝑥 = 0
𝜕𝑎𝑛 Ω Ω
 Trial solution procedures for differential equations
Weighted residuals sample problem
𝑑 𝑑𝑢 𝑥 2
• Consider the problem governed by the differential equation 𝑥 = 3 𝑥𝜖 ]1,2[
𝑑𝑥 𝑑𝑥 𝑥
𝑢 1 =2
With boundary conditions (BCs) 𝑑𝑢 1
−𝑥 =
𝑑𝑥 𝑥=2 2

Trial solution construction

• Construct trial solutions satisfying the BCs for all values of 𝒂. That is, whatever we choose as trial solution functions
must satisfy

𝑢 1; 𝐚 = 𝜙0 1 + 𝑎1 𝜙1 1 + 𝑎2 𝜙2 1 + … … … … + 𝑎𝑛 𝜙𝑛 1 = 2

𝑑𝑢 𝑑𝜙0 𝑑𝜙1 𝑑𝜙2 𝑑𝜙𝑛 1

−𝑥 = −𝑥 + 𝑎1 −𝑥 + 𝑎2 −𝑥 + … … + 𝑎𝑛 −𝑥 =
𝑑𝑥 𝑥=2
𝑑𝑥 𝑥=2
𝑑𝑥 𝑥=2
𝑑𝑥 𝑥=2
𝑑𝑥 𝑥=2
2
 Trial solution procedures for differential equations
Trial solution construction (cont)
• For example, choose a cubic trail solution function 𝑢 𝑥; 𝐚 = 𝑎1 + 𝑎2 𝑥 + 𝑎3 𝑥 2 + 𝑎4 𝑥 3
• At 𝑥 = 1 𝑢 1; 𝐚 = 𝑎1 + 𝑎2 + 𝑎3 + 𝑎4 = 2 (A)
𝑑𝑢
• At 𝑥 = 2 −𝑥 = −𝑥𝑎2 − 2𝑥 2 𝑎3 − 3𝑥 3 𝑎4
𝑑𝑥
𝑑𝑢 1
−𝑥 = −2𝑎2 − 8𝑎3 − 24𝑎4 = (B)
𝑑𝑥 𝑥=2
2

We have four coefficients and two constraints equations, so we can eliminate two coefficients
1
• Eliminate 𝑎2 from (B): 𝑎2 = − − 4𝑎3 − 12𝑎4
4
• Eliminate 𝑎1 from (A): 𝑎1 = 2 − 𝑎2 − 𝑎3 − 𝑎4
 Trial solution procedures for differential equations
Trial solution construction (cont)
• The expression of the trail solution becomes
1
𝑢 𝑥; 𝐚 = 2 − 𝑥 − 1 + 𝑎3 𝑥 − 1 𝑥 − 3 + 𝑎4 (𝑥 − 1) (𝑥 2 + 𝑥 − 11)
4
• We can rewrite the trail solution function as
𝑢 𝑥; 𝐛 = 𝜓0 𝑥 + 𝑏1 𝜓1 𝑥 + 𝑏2 𝜓2 𝑥
With
1
𝜓0 𝑥 = 2 − (𝑥 − 1) 𝜓1 𝑥 = (𝑥 − 1)(𝑥 − 3) 𝜓2 𝑥 = (𝑥 − 1)(𝑥 2 + 𝑥 − 11)
4
The residual is

𝑑 𝑑𝑢 𝑥 2 1 2
2
𝑟 𝑥; 𝐛 = 𝑥 − 2 = − + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2
𝑑𝑥 𝑑𝑥 𝑥 4 𝑥
 Trial solution procedures for differential equations
Point collocation
• With collocation points 𝑥1 = 4/3 and 𝑥2 = 5/3
Set 𝑟 𝑥; 𝐛 = 0
• The system of equations is

4/3𝑏1 + 4𝑏2 = 11/8

8/3𝑏1 + 13𝑏2 = 97/100
and the solution is
𝑏1 = 2.0993
𝑏2 = −0.3560
 Trial solution procedures for differential equations
Subdomain collocation
• With Δ𝑥1 = 1 < 𝑥 < 1.5 and Δ𝑥2 = 1.5 < 𝑥 < 2
.
Set Δ𝑥𝑖
𝑟 𝑥; 𝐛 𝑑𝑥 = 0

• The system of equations is

1.5
1 2
2 1 9 19
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 𝑑𝑥 = 0 𝑏1 + 𝑏2 =
1 4 𝑥 2 8 24
2
1 2
2 3 63 11
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 𝑑𝑥 = 0 𝑏1 + 𝑏2 =
1.5 4 𝑥 2 8 24

and the solution is 𝑏1 = 2.5417

𝑏2 = −0.4259
 Trial solution procedures for differential equations
Least squares
2 𝜕𝑟
• Set 1 𝑟 𝑥; 𝐛 𝑑𝑥 = 0
𝜕𝑏𝑖
𝜕𝑟 𝜕𝑟
• The derivatives of the residual are =4 𝑥−1 and = 3 3𝑥 2 − 4
𝜕𝑏1 𝜕𝑏2
• The system of equation is
2
1 2
2 16 7
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 4 𝑥 − 1 𝑑𝑥 = 0 𝑏1 + 27𝑏2 = 8𝑙𝑛2 −
1 4 𝑥 3 2
2
1 2 711 33
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 3 3𝑥 2 − 4 𝑑𝑥 = 0
2
27𝑏1 + 𝑏2 =
1 4 𝑥 5 4

and the solution is 𝑏1 = 2.3155 𝑏2 = −0.3816

 Trial solution procedures for differential equations
Galerkin 2
• Set 𝑟 𝑥; 𝐛 𝜓𝑖 (𝑥) 𝑑𝑥 = 0
1
• The system of equation is
2
1 2
2
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 (𝑥 − 1) 𝑥 − 3 𝑑𝑥 = 0
1 4 𝑥
2
1 2
− + 4 𝑥 − 1 𝑏1 + 3 3𝑥 − 4 𝑏2 − 2 (𝑥 − 1) 𝑥 2 + 𝑥 − 11 𝑑𝑥 = 0
2
1 4 𝑥

5 41 and the solution is 𝑏1 = 2.1378

− 𝑏1 − 𝑏2 = 29/6 − 8𝑙𝑛2
3 5
41 81 𝑏2 = −0.3477
− 𝑏1 − 𝑏2 = 211/16 − 24𝑙𝑛2
5 2
 Trial solution procedures for differential equations
Comparison
• The most precise solution is given by the Galerkin approach