Forum

Redesign Multiscale Structural Mechanics

  1. Wenbin Yu

    Dear All, I am writing to ask your suggestions for the Multiscale Structural Mechanics we have offered in the past several years. I have plan to redesign it so that it becomes more simulation based, less theoretical because I found some students are struggling with the theory. Any opinions are welcome. 

  2. Vivek Dhimole

    Dear Professor,

    It will be great for students and also please make the correction related some error issues which comes time to time when running dehomogenization in Ansys-Swiftcomp/Abaqus swift comp

    Thank you

  3. Nagappa Siddgonde

    Dear sir,

    We are genuinely facing the formulation of MSG theory due to its complexity to derive mathematical formulations and developing  MSG codes to verify results ourselves.  I kindly requested you to please write a book on MSG theory with detailed mathematical formulations, so that it would helpful for the research community. In fact, most of the time we used the MSG tool and predict the results but we are not able to understand the MSG theory what it is exactly and how? it computes.  So that we request you to please write a book on MSG theory along with Multiscale Structural Mechanics. Thank  you

    with regards

    Nagappa 

     

  4. Goon Mo Koo

    Dear Dr. Yu,

     

    Here is my opinion.

    I have learned a lot from this coursework. I liked the theory (especially, its computational power and easiness of use once I understand how to use the tool) and your explanation and derivation was straightforward to understand. The one thing I want to suggest is I hope we have some more explanation about tools as much as theory. Even though we understand the theory, it was difficult for me to apply it to software. If you plan to redesign this course more simulation based, I think it will be better you spend more time on explaining simulation work. In this coursework, what we had was a big picture about MSG in the course and you want us to see tutorial videos and documents. I think the problem we had for final project was simple and straightforward. So, maybe next year, you probably show how to do it in class step by step (and some explanation about how to figure out when some errors pop up) and assign them a bit more complicated project. In addition, while I am using the ECN computer, it was not easy to install SwiftComp plug-in. I think one of the points of this class is to learn how to apply theories to software, but we spent much time to install the software. I would recommend instruct the method of installation at the beginning of the class if you insist to use this tool or reserve a computer lab which has this software in all computers and ask students to work at that computer lab.

    Plus, I also agreed with Nagappa that I would also want to read a book which has more details in MSG. While I was reviewing the MSG, Chapter 7, Beam theory, and a document that you posted for MSG wasn't enough for me. Other that these, I think this coursework was well-designed and I enjoyed the coursework. It was honored for me to take this course with you and hope everyone has a wonderful christmas and happy new year! 

     

    Sincerely,

    Goon mo Koo   

  5. Wenbin Yu

    Goon Mo, Thanks a lot for your comments. I will consider this next time when I offer it. ---- Emailed forum response from wenbinyu@purdue.edu
  6. Wenbin Yu

    Dear Nagappa, Thanks a lot! ---- Emailed forum response from wenbinyu@purdue.edu
  7. Junyu Wang

    Dear Prof Yu,

     

    The first half of the current course are highly focused on mathematics background. In my opinion, this part can be covered with less details, so that it takes less time. At least for my case, these materials were all reviewing. I believe that just quickly going over this part in a review fashion and assign some homework focusing on fundamental understanding will well serve the purpose of this part of the course. (I think it is fine to cut the length of chapter 2-5 in half)

    The following part of homogenization theories is an essential of this course. So please spend more time on it, especially MSG. MSG was a main drive for me to learn this course. I think it would be great to enhance the understanding by doing a couple of examples problems, including a couple of simulation demo (Abaqus for example, also the demo can be done by PhD student :) ). 

    I think it would also be great to cover a little more on dehomogenization, especially using simulation software.

    Overall, spending more time on details of MSG and micromechanics would be a good direction, and including more examples for demonstration greatly helps. 

    I hope my suggestions help.

     

    Have a merry Christmas,

    Junyu

  8. Wenbin Yu

    Thanks a lot for the excellent suggestion. ---- Emailed forum response from wenbinyu@purdue.edu