Autumn 2021
Prof. Avijit Chatterjee
Author: Vishnu Sankar
Pre-requisite courses: None
Pre-requisite skills: Basic coding skill in Python/MATLAB is useful
Course Content: ASC course description is updated, kindly check it out. —
Coure Evaluation: 3 assignments - 80% weightage 1 quiz in place of endsem - 20% weightage
Motivation behind taking the course: This course is one of the very hands-on courses which makes us of programming languages to solve system of hyperbolic linear and nonlinear equations concerning Computational Fluid Mechanics through a Eulerian approach. I was interested to learn about CFD and hence it hooked me up :P
Information about Projects/Assignments: There are three computational assignments which contribute to 80% of the course evaluation. Slightly tricky assignment as it becomes messy to de-bug the code.
Overall Course Difficulty: 4/5
Average Time Commitment: The assignments can be very time consuming if you get a bug, which mostly you will get for sure :P. So each assignment can take about anywhere between 4-12 hours based on your skill to de-bug the code
Attendance Policy: None
General funda: Solve the assignments by yourself and pay attention to the results and physics of the plots, this is where the questions were for the final quiz. This quiz is the final test that decides your grade in this course.
Professor’s Teaching Style: Coherent and lucid lectures, sometimes with a philosophical tinge. There are no lecture slides, the professor follows a book religiously which can be read to brush up your concepts. Lecture videos are more than enough to get a grip on the concepts
Who can take this course?: Anyone with an interest to learn more about ways to computationally solve system of equations using Numerical Methods can take up this course. The course ends with a sneak peek into the modern algorithms used in building commercial software like ANSYS, it would be more fun to explore this area with more challenging problems. This area has been widely studied in the last 70 years so most of the techniques are already pretty developed. The current state of the art which is still in research stage is the DG method