User login

Navigation

You are here

""STUDY GROUP""

Mahdi Kazemzadeh's picture

This is a great news for all students here in iMechanica! The formation of a study group which will focus on the basic concepts and the facts of daily research, the simple tools that any of us will need at a certain stage. Please see blow the formation of this idea and add your comment, it would be good to know other points of view as we can improve the initial idea step by step.

Early on this year we started a new topic here which we thought will attract students and help us to build a community for students in iMechanica, that would have lot of advantages at this stage. You can go through the development of this idea at different stages at" calling all students ". Recently we have got a comment on our forum which follows, as I am thinking that this voice could have been representing a wide group of students and their ideas therefore I thought it is better if we "all students" post our suggestions and responds.  

This can be a further start point for us, if we know what the other students think. Recently, "Michael Anthony" has post a comment to our link which I do appreciate because of being honest and telling us what he thinks we should be doing here. I am trying to put this to your attention once again and keep this idea fresh. His comment was some sort a question from most of us as so called "Student Society" in iMechanica. I think it is important if we all think about this comment and respond of what we think is top! 

"Micheal" says: 

"I am commenting because you requested that students who read this comment. As to the topic at hand, I don't think much action is required. I do not think that the community here is going to make any large shifts because of a special effort to attract us students.  I do not think that a section called Students would be very effective. It is of note that the similar section you mentioned—Education—has many entries that do not seem to have anything to do with education.  

Forgive my cyncicism, but I am sharing what I think is most realistic. I don't think iMechanica needs to change to suit students; it has currently formed into what it is naturally and organically, and it doesn't make sense to change it to try to suit such a goal. Publicity among students is another issue, but I do not think it makes sense to try to make this a different place to attract people who otherwise would not be interested."

Lets get this opportunity to ask from ourselves; " how we can get more students join us here?" Lets ask from ourselves that why some students think that iMechanica is not helpful for them or in other words should not focus on that group. I am inviting all students and Micheal to discuss this further and refresh our topic, maybe to call students another time! I am quite sure if we start thinking about Micheal's comment we can find better ways to improve iMechanica to suit students. Lets take it from here and try to give practical comments and replies to a question which might be a query of other students.

Comments

I am a bit embarrassed to have resurrected such an old
topic (I did not notice how dormant it was when I posted my response. I
forget how I navigated to the page.), but I am glad if it refreshes
conversation that is important to the community here.

You pose the question why some students think that iMechanica is not
helpful for them. It seems to me like it would be very hard to convince
someone who isn't interested in a platform like iMechanica that it is
somewhere they want to spend time. I suspect most people who are
excited about mechanics would try to fit in here.

In my own experience, I regularly visit iMechanica since I've heard
about it, though I mainly lurk rather than post myself. It is somewhat
intimidating to be surrounded by so many accomplished mechanicians, but
that does not discourage my being present here, quite the opposite.

(I am the `Michael Anthony' spoken of above. Previously I used my full
name—Michael Anthony Graham—as my user name and it got cut off before
my surname.)

I am a student of USTC.

I am research in Biosensor base Microcantilver .

yoursdhruly's picture

Title borrowed from a song by the Who

 

Michael, Mahdi,
It's been a while, hasn't it? :-)

Here
are my thoughts. Last time around, we started a similar thread, that
was an interesting discussion but honestly (and bluntly) speaking, went nowhere. Let us do things differently this
time around. Let us use this thread to develop one or two concrete
ideas that we can work on WITHIN the framework of iMechanica. Let us
not ask for special pages or special considerations. After all, aren't
we all a student of something or the other?

 

Here are 2 quick suggestions that come to my mind right now:
1.
A journal Club - (Prof. Suo had suggested this on the first thread) My
suggestion is we do this differently from the existing jClub in that we
focus on key papers in the development of an area of research. While
the jClub focuses on "hot" topics, we can take the complete opposite
direction and focus on (relatively) "established" topics. Ductile
fracture, for example. Or thin films. And I argue we can do better than
the jClub - we have a MUCH larger audience than they do, since everyone
is a student, right? :-)

 

2.
More Student oriented posts: For example - the eternal academia vs
industry conundrum, or a "best textbook" recommendation. With a little
creativity, we can encourage more student activity.

 

In the
end, we must ask ourselves one very important question: WHY? To me the
answer is two-fold: While the established mechanician today may be
difficult to get involved, the established mechanician of tomorrow is a
student today. Secondly, I think this is a genuine gap in iMechanica,
perhaps in the mechanics community in general, but I do not have the
experience to say that with any confidence. Interaction is ofetn
reluctant and I am willing to guess collaboration is not as prolific as
in other branches (life sciences for example). The reasons for this may
be disparate (cultural, nature of work, insecurity). The point is,
iMechanica represents a shift away from this insulated existence. And
like several famous and not-so-famous people said, catch 'em young.

 

I am looking
forward to this gathering critical mass in the coming days and bearing
fruit in the form of something concrete. To me the most promising is
the "student jClub" or "Online Study Group" - OSG. Let us define if
this is indeed what we want to do, define the scope and start a new
thread to invite everyone to define the form and structure, similar to
the jClub.

 

Michael, thanks for rekindling these thoughts. And Mahdi, thanks for bearing the torch, as always!

Zhigang Suo's picture

Dear Dhruv:  Congratulations on your new job at Intel!  Now you count as an industrial iMechanician.  Your idea of a Study Group focusing on relatively established topics sounds very attractive.  It would be great you and others can refine this idea and result in a working plan.

The idea of a study group is very interesting.  As a perpetual student, I would like to know about a problem that's been bothering me for a while - stress singularities.

We know that stress singularities are a feature of linear elasticity.  Following Williams' solution that showed that stress singularities occur at the corner of a wedge under certain boundary conditions, a huge amount of work has been done in quantifying such singularities for bimaterial and multimaterial interfaces.  It has been found (see the work of Theocaris etc.) that multimaterial interfaces where two or more materials met at a point almost always lead to stress singularities.

However, stress singularities imply that the stress is unbounded - so for a non zero strain the stress immediately jumps to infinity.  Cleraly, such a thing cannot be sustained by real materials.  So, how do real materials deal with singularities at trimaterial interfaces?  Also, how should we model such situations numerically?

I hope some of you can shed some light on that matter. 

 -- Biswajit

Zhigang Suo's picture

  1. The best known case of a singluar stress field is around a crack.  A way to use the singularity was developed by Irwin and others.  The subject is of course fracture mechanics.  Here is a short review of Irwin's approach by Bao and me.
  2. Irwin's approach has been extended to deal with sigular stress field in other situations, such as a wedge, and bimaterial corner.  This extension has been developed by several groups.  Here is a recent paper from my group.  The paper contains a review of the literature.
  3. There are many ways to treat the singular stress field using the finite element method.   A few examples follow
  • Remove the singularity by prescribing a traction-displacement relation.  See an iMechanica thread initiated by Dhruv Bhate.
  • Obtain the singular stress field by using a fine mesh and extrapolation.   My students have been using this method.
  • Use conservative integrals to obtain the  multiplication factor of the  singular  field.  For example, use the J-integral to calcuate the stress intensity factor of a crack.
  • Embed the konwn functional form of the singular field into the finite element method, called the Extended Finite Element Method.  There are quite a few experts on this method among iMechanica users.
Rui Huang's picture

As Biswajit pointed out, stress singularity is a feature of linear elasticity, physically incorrect but yet practically useful. Linear elastic fracture mechanics is an example of practical applications of stress singularity. In practice, one may handle stress singularity numerically or physically. Various numerical methods have been developed to model stress singularity such as those mentioned by Zhigang (J-integral, XFEM, etc.). The physics or mechanics remains unchanged. As long as you are comfortable with the singular field as stated by small-scale yielding in linear elastic fracture mechanics, the problem can be readily handled in most cases.

Alternatively, various physical models have also been considered to remove the unrealistic singularity, typically by incorporating physical ingradients other than linear elasticity. The concept of crack bridging as reviewed by Suo and Bao or cohesive zone modeling in the similar spirit takes into account of the nonlinear interactions around the crack tip as the crack opens and grows. Other approaches have included micromechanics of damage and plasticity (strain-gradient plasticity, discrete dislocations, etc), and more recently, multiscale approaches are proposed with different levels of details (from continuum mechanics to first-principle quantum mechanics). Many questions remain open in this direction. Which approach to use would largely depend on the problem specific purposes as well as technical resources available.   

RH

Mahdi Kazemzadeh's picture

Hi Dhruv

I found very nice ideas in your comment, they can be a good start point. My suggestion is that we give this post some more time. I think first of all we should get more students to read this post. We will start working on your ideas step by step. Thanks for brilliant ideas, also congratulations for the new job!

Lets keep this interesting discussion going.
 

Zhigang Suo's picture

I've thought more about Dhruv Bhate's idea of Study Group.  His idea is appealing: 

the mission of the Study Group can be simply to discuss ideas of lasting value to mechanics. 

We learn difficult ideas by discussing and by teaching one another.  We appreciate these ideas better by relating them to other ideas, and to problems of current interest.  Some sample ideas can be

The Study Group can run like the jClub, with an Editor recruiting Discusiion leaders, each Discussion Leader initiates a Theme, and users discuss.  As Biswajit Banerjee pointed out, many of us are perpetual students.  The Discussion Leaders can be recruited among students of all ages.

I'd love to hear from you about this and related ideas. 

Xuanhe Zhao's picture

As a student and daily user of iMechanica, I find the idea on Study Group quite attractive. I would like to add one more suggestion. Besides general ideas, the study group may also consider some specific technical questions, sush as "how to solve a type of boundary value problems with ABAQUS". As we can see, many technical problems raised by students on imechanica are relatively unattended. Examples include:

For engineers and engineering students, the answer to these questions is of essential importance to their current projects. A wise clue may save them months of struggle in mud. Therefore, a study group focused on technical questions would attract more attention from student side.

Anurag Gupta's picture

Prof. Suo,

 I find the idea of such a group very useful. It can differ from (and thereby enriching it) the contents of jclub, by laying more stress on the fundamental ideas onto which solid mechanics rests. Our discussions can range from classroom problems to classical works. 

Anurag

Teng Li's picture

The proposal of Study Group with focus on ideas of lasting values are of general interest, especially among student users in iMechanica.  It can be supplemental to the jClub, whose focus is on more frontier research fields.  While being constantly successful in introducing exiting topical research fields to general audience, the jClub often leads to discussions requiring particular expertise to follow and participate. The ideas of lasting values such as those Zhigang pointed out, as well as the pratical problems in doing everyday research as pointed by Xuanhe, are worth of discussions in a platform like iMechanica.  If successful, such discussions in long run may paly a vital role in achieving the iMechanica mission:  "Pave a way to evolve all knowledge of mechanics online".

Mahdi Kazemzadeh's picture

I am very pleased to see that our initial ideas are getting formed and has found a perfect shape now; "study group". The idea of study group is very much interesting and I am pretty sure that would make many students to join us as they will see we have focused on some basic concepts and are trying to learn them more in depth. I think the form of this can be very similar to the jclub just with a slight changes as we would be more focused on concepts than every day research, well we might find some case examples from ongoing research as well to mention there.

I dont see any particular issue with doing this, I have been looking at the links that Prof. Suo has mentioned in his comment up here and trying to find the best path to start. I suppose we can advertise that we are starting our "study group" and start it as a fresh idea over here. At this stage I think our "study group" dissolve some new ideas that are mentioned in comments here, we can always add new things to it. Personally I will do as much as I can to assist.

The first step is to put the initial idea in more practical form, as if we think it is fine to have it similar to current jclub then it is quite easy to start. I myself found the style of jclub pretty good and well designed, but if you have different ideas, please let me know. I try to put this post up once agian just to make other students aware of our new plans. All the best.

 

Mahdi

Temesgen Markos's picture

This is a very good initiative. After doing course work (mostly from books), graduate students move to reading "state of the art" papers and there is little chance of reading the classic works as originally written. May be this can be an opportunity to do so if we can spare some time to that. Here is a good example of what Prof. Trefthen from Oxford has done with applied math: 13 classic papers in applied mathematics

 The senior members of the forum can help by selecting classics in the different branches as Prof. Suo has done here. We can have five or more papers on a given topic starting with the earlier works and progressing to the recent. We can take topics like fracture, stability, soft materials, finite elements, etc. Let's first make a list of topics that can be discussed and the time to be spent.

I suggest a month for a given topic. I agree with Prof. Suo's suggestion that discussion leaders be recruited among student. Probably it is also good to select one senior member as an advisor to the study group.   

 

yoursdhruly's picture

That is a good idea, Temesgen! I agree we should have an overseeing advisor who can review the discussion and comment on any errors or digressions. I personally think 1 month is too short of a time, 2 months would perhaps be better for the study group to reach a meaningful conclusion for each of its topics.

yoursdhruly's picture

This may be a slight
digression, but I hope you still find it relevant to the "Study Group".
Firstly, it is wonderful to see this get some traction.

 

I
was thinking of the Study Group idea this morning, and I said to
myself, we should take it a step further. Let us write a review
article! Not particularly with the intention of submitting it to a
journal, but merely to collect it one document that we can then make
freely available on iMechanica. Of course, this would require a
significant contribution from the study group leader and other people, but on the
other hand, since this would be a review article, anybody could
contribute bits and pieces. At the end, we can all list ourselves as
authors with the order of listing awarded according to level of
contribution. Hows that for democratic online collaboration? Smile

 

So what topics could
we tackle? I am biased in my thinking, according to my experience and
education, but feel I should put it out there regardless, in the hope
that it will stimulate some thoughts in your mind. The review article
must be in an area where enough of us can make useful contributions and
where there would be an interest in such a review paper. I think the
interface of academic theory/experiments and industrial application
is one such area, e.g. fracture mechanics in microelectronics packaging.
Another is some key mysterious areas such as stress singularities,
discussed above.

 

This will give our Study Group some
direction and we will emerge with a living document that will benefit
all who participated and all who follow. It may even be an incentive to
encourage greater participation. I am not sure how realistic such an
effort is, and would appreciate any comments.

Mahdi Kazemzadeh's picture

Dhruv

Could you please explain just a bit more about your idea of "review paper", If you mean having some sort of review system, it seems it goes further than study group, I am not sure?! I will be happy to read your idea frist about how we should establish the study group first! I think we have to focus on just software bit of that! Like what we should start with and how, It would be much easier if we develope it step by step and see how it goes.

 

Mahdi

yoursdhruly's picture

Mahdi,
I do agree with you that we are still a long ways from formalizing the study group itself and just wanted to put this idea out there. However, I think you misunderstood what I meant about the review article. It was not meant to be a system, only a way to document the discussions that we have in the study group for future reference. I hope that makes sense.

Also, do you know why this was taken off as a "front page sticky"? Is there anyway to get it back? Or should we just start a new thread?

Thanks,

Dhruv 

Mahdi Kazemzadeh's picture

Dhruv

Thanks, it makes sense now. We will certainly need something to document the discussions that we are going to have there in the study group. I do agree with you that we still have a long way to go! I think there is no problem if this takes long to form, it will help other members to keep track of what we are doing. At this moment the best thing to do is to create a platform for our idea, I suppose the style of jClub as start point will help to form the initial idea. Maybe you could comment on this, I mean the first step. The jClub style and form is quite representing and good so we could just review that for our purpose but I would like to know your idea as well.( I have also read some where in your comments about this but it would be better if we just could build the initail idea and take it from here). 

I posted this on front page as "sticky post" and contacted Prof. Suo for how long we can have the post on front page, It was there for few days. I think because there are already 2 sticky posts up there, this was taken off. It had more than 1000 reads just in two days which is brilliant! I think later on after we built up the study group we can have one headline reserved for our study group at sticky posts location! Thanks.

Mahdi

Zhigang Suo's picture

yoursdhruly's picture

Prof Suo,
Thanks for the clarification, your argument makes sense. Thanks also for the link on the "what we talked about" section.

I think the operating notes of the jClub is the best place to start. I will post an edited version  for the study group soon, though I wonder if I should do so in a new thread to get more participation. 

Mahdi Kazemzadeh's picture

Dhruv

I am quite happy that we have the same style of jClub for the study group. I think if you want we can divide all the work between eachother at this time. So we both can keep it going, just let me know if we should draft something first or you have some other ideas to start with. 

Mahdi

Subscribe to Comments for """STUDY GROUP"""

More comments

Syndicate

Subscribe to Syndicate