ES 622: Finite Element Methods - Spring 2022

Table of Contents

Basic Information

Class timings Tuesday, Wednesday, Friday, 11:05-12:00 hrs
Class location AB 7/101
Online link Class link / Live Notes
Instructor Gaurav Srivastava (gauravs@iitgn.ac.in)
TA Dravesh Yadav (dravesh.yadav@iitgn.ac.in)

Course objectives, syllabus, books, pre-requisites

General philosophy of the course

In a world where a large number of simulation tools are available and most people assume they know how to perform a finite element analysis, this course focuses on the fundamentals and aims to equip the students with the 'how' and 'why' of finite element simulations. It covers the fundamental aspects of the finite element method that are required to understand what goes on under the hood in available simulation software.

Objectives

  • To learn the fundamentals of the finite element method.
  • To develop a computer program to perform finite element analysis of simple PDEs.
  • To develop an understanding of what goes on under the hood of commercial finite element packages.

After finishing this course, you will

  • know an alternative representation of physical laws (called weak form)
  • be able to convert any given linear partial differential equation (PDE) to its corresponding weak form
  • be able to develop finite element formulation for any physical problem governed by a PDE
  • have developed a computer code for solving linear PDEs using the finite element method
  • know about different types of finite elements and their suitability for different types of physical problems
  • have some idea of potential problems that can arise in typical finite element solutions
  • know how much error can be expected in finite element analysis

Syllabus

  • Review of basic mathematical preliminaries.
  • Strong and weak forms, Galerkin's approximation.
  • 1D, 2D, and 3D finite element formulations, isoparametric elements. Error behavior.
  • Finite element formulation of elliptic PDEs (elasticity equation), parabolic PDEs (heat equation), and hyperbolic PDEs (wave equation).

Reference Books

  • An Introduction to the Finite Element Method - J.N. Reddy.
  • Introductory Finite Element Method - C.S. Desai.
  • Finite Element Procedures - K.-J. Bathe.
  • The Finite Element Method: Linear Static and Dynamic Finite Element Analysis - T.J.R Hughes.

Pre-requisites

  • Knowledge of basic linear algebra:
    • rank, column space, null space of a matrix,
      • solving system of linear algebraic equations,
      • computing eigenvalues and eigenvectors.
  • Knowledge of a programming language:
    • Matlab is a good option. Here is an excellent tutorial.
    • Scilab is an open source alternative to Matlab.
  • In case you feel adventurous, you can use C++ or FORTRAN as well.

Course Policies

Etiquette

  • Please be considerate about everyone's time.
  • In all emails pertaining to this course, please have "ES622" in the subject line.
    • (note that there is no space or hyphen or anything between ES and 622)

Cheating

Cheating cases (assignments/codes/exams/project) will be awarded an F grade and will be reported to academic office. It is expected that this will never happen and everyone will uphold the honor code.

Late submissions

All delays beyond the defined deadlines will attract reduction in marks as per the following curve. The reduction factor, \(R\) will be multiplied to the obtained marks. Mathematically, it is given by: \(R = \exp(-D^2)\), where \(D\) is the total delay in days (will be counted hourly, i.e. fractional days are possible). deadline-delay-reduction.png

Grading

Following will be the weightage of different components of assessment

Component Weightage
Homework assignments 20%
Spot quizzes 15%
Exams (mid, end) 25% each
Project 15%

Modes of formal assessment

  • There will be two types of assignments: analytical/hand calculation and coding.
  • For coding assignments, submission of source code will be required.
  • Expect one assignment per week.
  • Spot quizzes will primarily be objective type. Expect one quiz per week.
  • Exams will primarily be subjective / coding type. They may be in-class or take-home.

Emphasis on self-learning

It is important to develop the habit of self-learning. A number of reading assignments and self-exercises will be given during the course. These will not be formally graded and it will be expected that students will go through them on a regular basis on their own.

Project

The basic idea of a project is to utilize the knowledge gained from this course to a real-life situation or to better understand certain concepts that remain hidden otherwise. Try to answer some/more such questions during the project. Ideally, you would choose a reasonably complicated looking real-world problem and analyze it using a software (either the one you will develop as part of the course, or ANSYS, or ABAQUS, or any other that you may know of).

Timeline

Project is to be done in groups of not more than 3. Following timeline must be adhered to for all submissions. (this timeline will be updated during the first week of classes)

Date Task Marks
Feb 4 Formation of groups, identification of topic. SUBMIT group details and abstract of proposed work. 10
Feb 18 Feedback from instructor about project topics -10*
Mar 17 SUBMIT 2-3 page detailed report discussing overall approach, idealizations, etc. required for project 20
Apr 16 Project presentations (one per group) 30
Apr 18 SUBMIT final report. 40

\(^*\) In case the instructor delays in giving feedback, every group gets 10 bonus points.

Guidelines on intermediate detailed report

This should have a clear roadmap of the activities being planned including the problem definition, governing equations, solution methodology, objectives of the study, and final deliverables.

Guidelines on the final report

This report should be organized as follows:

  1. Introduction (including background and motivation for the chosen problem)
  2. Literature Review (a brief review of recent literature dealing with the chosen problem)
  3. Theoretical Formulation (details of governing equations - PDE and weak/energy form, discretization)
  4. Verification/validation (to demonstrate the accuracy of the solution approach)
  5. Numerical studies (e.g. to show effects of parameters, discussions on physical behavior)
  6. Conclusions (summary of main findings of the study)
  7. References

Following are the titles of some projects done in previous years

  • Free Vibration Of Thin Plates
  • Finite Element Modelling of Thermal Management Systems of Laptops for effectiveness analysis
  • Analysis of Cold Rolling process using Finite Element Analysis in Ansys Workbench
  • Static Analysis of Leaf and Coil Spring
  • Design and Analysis of Rolling Process
  • Thermal expansion and Stress analysis of the Radial Turbine
  • Modelling Reaction in Batch Reactor

Calendar (tentative)

Lec No Date Topic Attachments
1 4 January, Tuesday Stretching of 1D bar. Derivation of governing equations - force equilibrium, energy and virtual work forms. V1 / V2
2 5 January, Wednesday Notion and equivalence of strong and weak forms. Fundamental lemma of calculus of variations V3 / V4 / V5
3 7 January, Friday Euler-Lagrange equation. Ritz-Galerkin method of solving weak form V6 / A1 (due 13 Jan)
4 11 January, Tuesday Ritz-Galerkin method of solving weak form V7
5 12 January, Wednesday Idea of spatial discretization. Application of boundary conditions. Formalization of FEM. V8
  14 January, Friday Makar Sankranti A2 (due 21 Jan)
6 18 January, Tuesday Element-level view of formulating FEM matrices. Boundary conditions. Quiz 1 / V9 / V10
7 19 January, Wednesday Properties of shape functions. Formulation of 1D reference element. V11 / V12
8 21 January, Friday Gauss quadrature. Computer implementation of 1D FEM. A3 (due 28 Jan) / V13
9 25 January, Tuesday Computer implementation of 1D FEM. Quiz 2
  26 January, Wednesday Republic Day  
10 28 January, Friday Construction of 1D shape functions using Lagrange polynomials. Error computations. A4 (due 4 Feb)
11 1 February, Tuesday Computing weak form of 2D Poisson equation. Quiz 3
12 2 February, Wednesday 2D FE formulation.  
13 4 February, Friday Isoparametric bilinear element.  
14 8 February, Tuesday Isoparametric bilinear element. Quiz 4
15 9 February, Wednesday Treatment of boundary integral in 2D problems.  
16 11 February, Friday Treatment of boundary integral in 2D problems.  
17 15 February, Tuesday Consideration of vector fields in 2D. Quiz 5
18 16 February, Wednesday Consideration of vector fields in 2D.  
19 18 February, Friday Weak form of time dependent problems. A5 (due 4 Mar)
20 22 February, Tuesday Weak form of time dependent problems. Quiz 6
21 23 February, Wednesday Time integration of parabolic problems.  
22 25 February, Friday Time integration of hyperbolic problems.  
  26 February - 5 March Mid Semester Exam Week  
  27 February, 9-11 AM Mid Semester Exam (7/210)  
  6 March - 13 March Mid Semester Recess Week  
23 15 March, Tuesday Time integration.  
24 16 March, Wednesday 1D functionals.  
  18 March, Friday Holi  
25 22 March, Tuesday 1D functionals and their link with strong and weak forms.  
26 23 March, Wednesday 1D functionals with multiple arguments.  
27 25 March, Friday 1D functionals.  
28 29 March, Tuesday 2D functionals. A6 (due 5 April)
29 30 March, Wednesday 2D functionals.  
30 1 April, Friday Locking behaviour.  
31 5 April, Tuesday Issues in usual FEM in incompressible materials. A7 (due 12 April); Quiz 7
32 6 April, Wednesday B-bar method for handling incompressible problems.  
33 8 April, Friday Hu-Washizu variational principle.  
34 12 April, Tuesday Shear locking in elasticity problems.  
35 13 April, Wednesday Shear locking in elasticity problems.  
  15 April, Friday Good Friday  
36 19 April, Tuesday Numerical artifacts in different FE elements.  
37 20 April, Wednesday Method of manufactured solutions for verification of numerical methods.  
38 22 April, Friday Calculation of secondary quantities like stress, strain, and flux.  
  26 April - 4 May End Semester Exam Week  
  2 May, 2-4:30 PM End Semester Exam (7/103)