Vaclav, Bruno Thanks to both of you very much. Then how about the convertion from the stress to the force on each particle, where the stresses are from the boundary ?
kan On 1/23/08, [EMAIL PROTECTED] < [EMAIL PROTECTED]> wrote: > > Send Yade-users mailing list submissions to > [email protected] > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.berlios.de/mailman/listinfo/yade-users > or, via email, send a message with subject or body 'help' to > [EMAIL PROTECTED] > > You can reach the person managing the list at > [EMAIL PROTECTED] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Yade-users digest..." > > > Today's Topics: > > 1. Re: How to get the micro-force from the macro-stress (kan) > 2. Re: How to get the micro-force from the macro-stress > (Bruno Chareyre) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 22 Jan 2008 23:58:21 -0600 > From: kan <[EMAIL PROTECTED]> > Subject: Re: [Yade-users] How to get the micro-force from the > macro-stress > To: [email protected] > Message-ID: > <[EMAIL PROTECTED]> > Content-Type: text/plain; charset="utf-8" > > Thank you, Vaclav, > > So the step for this is : > during each timestep, the position of each particle/element keeps the > same, so we can calculate the force brom the force-displacement > relationship(in this case, the order does not affect the calculation since > all the particle/elements do not move), after we get the forces, we use > this > forces to calculate the acceleration, and then the velocity(average > velocity), and then the movement/displacement in this timestep, and then > update the position of each particle/element to new positions by add > movement/displacement to current position. (until now, the calculation > cycle > of this timestep is completed) > then go to next timestep by use the new positions of particles/elements as > the current position to do the next calculation cycle. > > is my understanding right? > > > > Thanks again. > > Kan > > > > On 1/22/08, [EMAIL PROTECTED] > <[EMAIL PROTECTED]> > wrote: > > > > Send Yade-users mailing list submissions to > > [email protected] > > > > To subscribe or unsubscribe via the World Wide Web, visit > > https://lists.berlios.de/mailman/listinfo/yade-users > > or, via email, send a message with subject or body 'help' to > > [EMAIL PROTECTED] > > > > You can reach the person managing the list at > > [EMAIL PROTECTED] > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Yade-users digest..." > > > > > > Today's Topics: > > > > 1. How to get the micro-force from the macro-stress (kan) > > 2. Re: How to get the micro-force from the macro-stress > > (V?clav ?milauer) > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Mon, 21 Jan 2008 13:29:54 -0600 > > From: kan <[EMAIL PROTECTED]> > > Subject: [Yade-users] How to get the micro-force from the macro-stress > > To: [email protected] > > Message-ID: > > <[EMAIL PROTECTED]> > > Content-Type: text/plain; charset="utf-8" > > > > Hi, buddies, > > > > I have some questiones. > > 1. how to get the micro-force from the macro-stress? > > As we know,for most of current studies, the concept of stress is used to > > describe the stress status. but when we look into the micro level, we > have > > > > to convert the macro-stress to descrete force that applied at each > > particle, > > then how to do this convert? > > > > 2. what is the calculation process in each time step? > > The problem is that, as see the relation ship below, > > ___\|/____\|/___\|/________ > > 01---02---03---04---05---06... > > | \ | / | | | | ... > > 11---12---13---14---15---16... > > | | | / | \ | |... > > 21---22---23---24---25---26... > > ... > > ... > > x1---x2---x3---x4---x5---x6... > > > > in which, the number is the ID of a particle, > > and "--- " , "/" ,"\" and "|" between the numbers mean there is a > contact > > between the particles > > "\|/" means force. > > assume the 01-02-03-04...is the upper bounday layer, and the force is > > applied as shown "\|/". > > when the calculation begins, we need to detecte the contact and update > the > > > > position of each particles, > > now my problem comes: > > what is the procedure to calculation and update the particles' > > information, > > you know the force is from up layer, so the first layer > ("01-02-03-04...") > > > > should be calculated, then the second layer,and then the third layer.... > > but > > in DEM, the particle is independent to move, so after some time, some > > particles in the first layer may move to the second layer and some > > particles > > in second layer may move to the first layer(other other layer),then how > > should I do in order to know which particle should be calculated and > > updated > > first and which should be done later in the same timestep? > > For example, > > 01 has contacting neighbors 02 and 11, in a same timestep, after we have > > calculated 01's properties( velocity, position....), we should calculate > > 02 > > first, then 03(the neighbor of 02), after the layer of 01-02-03...is > done, > > > > then we can step to calculate the 11, which is to say, in a same > timestep, > > though 02 and 11 are both contacting neighbors of 01, we should > calculate > > 02, and 11 later though 02 and 11 are both affected because of teh > > movement > > of 01. so, how to indentify which particle should be calculated and > update > > first , and which should be later? > > (because it is a dynamic process, so just use the order of ID is > not a > > way to solve this) > > > > Hope I have described the problem 2 clearly, > > > > Need your help, thanks so much. > > > > Kan > > -------------- next part -------------- > > An HTML attachment was scrubbed... > > URL: > > > https://lists.berlios.de/pipermail/yade-users/attachments/20080121/93495eff/attachment-0001.html > > > > ------------------------------ > > > > Message: 2 > > Date: Tue, 22 Jan 2008 09:20:53 +0100 > > From: V?clav ?milauer < [EMAIL PROTECTED]> > > Subject: Re: [Yade-users] How to get the micro-force from the > > macro-stress > > To: [email protected] > > Message-ID: <[EMAIL PROTECTED]> > > Content-Type: text/plain; charset=UTF-8 > > > > > > > so, how to indentify which particle should be calculated and update > > > first , and which should be later? > > > (because it is a dynamic process, so just use the order of ID is > > > not a way to solve this) > > The trick is that you have constant positions during the whole timestep, > > you calculate forces from all contacts (order doesn't matter, elements > > don't move during calculation) and then, before next timestep begins, > > there is LeapFrogPositionIntegrator, that calculates displacements based > > on velocity, inertia, applied forces, ... and moves elements to the new > > position. Obviously, your timestep must be small enough so that you get > > convergence. > > > > HTH, Vaclav > > > > > > > > ------------------------------ > > > > _______________________________________________ > > Yade-users mailing list > > [email protected] > > https://lists.berlios.de/mailman/listinfo/yade-users > > > > > > End of Yade-users Digest, Vol 19, Issue 14 > > ****************************************** > > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > https://lists.berlios.de/pipermail/yade-users/attachments/20080122/219e57e1/attachment-0001.html > > ------------------------------ > > Message: 2 > Date: Wed, 23 Jan 2008 10:29:14 +0100 > From: Bruno Chareyre <[EMAIL PROTECTED]> > Subject: Re: [Yade-users] How to get the micro-force from the > macro-stress > To: [email protected] > Message-ID: <[EMAIL PROTECTED]> > Content-Type: text/plain; charset=UTF-8; format=flowed > > kan a ?crit : > > Thank you, Vaclav, > > > > So the step for this is : > > during each timestep, the position of each particle/element keeps the > > same, so we can calculate the force brom the force-displacement > > relationship(in this case, the order does not affect the calculation > > since all the particle/elements do not move), after we get the forces, > > we use this forces to calculate the acceleration, and then the > > velocity(average velocity), and then the movement/displacement in > > this timestep, and then update the position of each particle/element > > to new positions by add movement/displacement to current position. > > (until now, the calculation cycle of this timestep is completed) > > then go to next timestep by use the new positions of > > particles/elements as the current position to do the next calculation > > cycle. > > > > is my understanding right? > > Exactly right. What you describe here is the centered second order > finite diffence scheme (also called leap-frog), which is used in Yade. > It is probably the most common (and most simple) scheme in DEM codes. > Other schemes may be possible but are not implemented. > > Bruno > > > > > > > > > Thanks again. > > > > Kan > > > > > > > > On 1/22/08, [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]>* < > > [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]>> wrote: > > > > Send Yade-users mailing list submissions to > > [email protected] > > <mailto:[email protected]> > > > > To subscribe or unsubscribe via the World Wide Web, visit > > https://lists.berlios.de/mailman/listinfo/yade-users > > or, via email, send a message with subject or body 'help' to > > [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]> > > > > You can reach the person managing the list at > > [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]> > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Yade-users digest..." > > > > > > Today's Topics: > > > > 1. How to get the micro-force from the macro-stress (kan) > > 2. Re: How to get the micro-force from the macro-stress > > (V?clav ?milauer) > > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Mon, 21 Jan 2008 13:29:54 -0600 > > From: kan <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> > > Subject: [Yade-users] How to get the micro-force from the > > macro-stress > > To: [email protected] <mailto:[email protected]> > > Message-ID: > > < > > [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED] > >> > > Content-Type: text/plain; charset="utf-8" > > > > Hi, buddies, > > > > I have some questiones. > > 1. how to get the micro-force from the macro-stress? > > As we know,for most of current studies, the concept of stress is > > used to > > describe the stress status. but when we look into the micro level, > > we have > > to convert the macro-stress to descrete force that applied at each > > particle, > > then how to do this convert? > > > > 2. what is the calculation process in each time step? > > The problem is that, as see the relation ship below, > > ___\|/____\|/___\|/________ > > 01---02---03---04---05---06... > > | \ | / | | | | ... > > 11---12---13---14---15---16... > > | | | / | \ | |... > > 21---22---23---24---25---26... > > ... > > ... > > x1---x2---x3---x4---x5---x6... > > > > in which, the number is the ID of a particle, > > and "--- " , "/" ,"\" and "|" between the numbers mean there is a > > contact > > between the particles > > "\|/" means force. > > assume the 01-02-03-04...is the upper bounday layer, and the force > is > > applied as shown "\|/". > > when the calculation begins, we need to detecte the contact and > > update the > > position of each particles, > > now my problem comes: > > what is the procedure to calculation and update the particles' > > information, > > you know the force is from up layer, so the first layer > > ("01-02-03-04...") > > should be calculated, then the second layer,and then the third > > layer.... but > > in DEM, the particle is independent to move, so after some time, > some > > particles in the first layer may move to the second layer and some > > particles > > in second layer may move to the first layer(other other > > layer),then how > > should I do in order to know which particle should be calculated > > and updated > > first and which should be done later in the same timestep? > > For example, > > 01 has contacting neighbors 02 and 11, in a same timestep, after > > we have > > calculated 01's properties( velocity, position....), we should > > calculate 02 > > first, then 03(the neighbor of 02), after the layer of > > 01-02-03...is done, > > then we can step to calculate the 11, which is to say, in a same > > timestep, > > though 02 and 11 are both contacting neighbors of 01, we should > > calculate > > 02, and 11 later though 02 and 11 are both affected because of teh > > movement > > of 01. so, how to indentify which particle should be calculated > > and update > > first , and which should be later? > > (because it is a dynamic process, so just use the order of ID is > > not a > > way to solve this) > > > > Hope I have described the problem 2 clearly, > > > > Need your help, thanks so much. > > > > Kan > > -------------- next part -------------- > > An HTML attachment was scrubbed... > > URL: > > > https://lists.berlios.de/pipermail/yade-users/attachments/20080121/93495eff/attachment-0001.html > > > > ------------------------------ > > > > Message: 2 > > Date: Tue, 22 Jan 2008 09:20:53 +0100 > > From: V?clav ?milauer < [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> > > Subject: Re: [Yade-users] How to get the micro-force from the > > macro-stress > > To: [email protected] <mailto:[email protected]> > > Message-ID: < [EMAIL PROTECTED] > > <mailto:[EMAIL PROTECTED]>> > > Content-Type: text/plain; charset=UTF-8 > > > > > > > so, how to indentify which particle should be calculated and > update > > > first , and which should be later? > > > (because it is a dynamic process, so just use the order of ID is > > > not a way to solve this) > > The trick is that you have constant positions during the whole > > timestep, > > you calculate forces from all contacts (order doesn't matter, > elements > > don't move during calculation) and then, before next timestep > begins, > > there is LeapFrogPositionIntegrator, that calculates displacements > > based > > on velocity, inertia, applied forces, ... and moves elements to > > the new > > position. Obviously, your timestep must be small enough so that > > you get > > convergence. > > > > HTH, Vaclav > > > > > > > > ------------------------------ > > > > _______________________________________________ > > Yade-users mailing list > > [email protected] <mailto:[email protected]> > > https://lists.berlios.de/mailman/listinfo/yade-users > > <https://lists.berlios.de/mailman/listinfo/yade-users> > > > > > > End of Yade-users Digest, Vol 19, Issue 14 > > ****************************************** > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Yade-users mailing list > > [email protected] > > https://lists.berlios.de/mailman/listinfo/yade-users > > > > > -- > > _______________ > Chareyre Bruno > Maitre de conference > > Institut National Polytechnique de Grenoble > Laboratoire 3S (Soils Solids Structures) - bureau E145 > BP 53 - 38041, Grenoble cedex 9 - France > T?l : 33 4 56 52 86 21 > Fax : 33 4 76 82 70 43 > ________________ > > > > ------------------------------ > > _______________________________________________ > Yade-users mailing list > [email protected] > https://lists.berlios.de/mailman/listinfo/yade-users > > > End of Yade-users Digest, Vol 19, Issue 15 > ****************************************** >
_______________________________________________ Yade-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/yade-users
