iMechanica - Comments for "2. Is a mesh required in meshfree methods?"
https://imechanica.org/node/490
Comments for "2. Is a mesh required in meshfree methods?"enWhich error estimator is useful?
https://imechanica.org/comment/11415#comment-11415
<a id="comment-11415"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>
Dear Prof. Sukumar
</p>
<p>
I've applied Voronoi diagram for refinement process in which it's functional contains residuals of domain and boundaries. Which error estimator can be useful in adaptive refinement of meshfree methods?
</p>
<p>
best regards,
</p>
<p>
Jafar
</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Thu, 02 Jul 2009 13:21:55 +0000Jafarcomment 11415 at https://imechanica.orgThanks Sir
https://imechanica.org/comment/11373#comment-11373
<a id="comment-11373"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>Thanks Sir this is goog informatiom.</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Sun, 28 Jun 2009 04:25:34 +0000hiader k. mahbescomment 11373 at https://imechanica.orgBackground grid for integration
https://imechanica.org/comment/2253#comment-2253
<a id="comment-2253"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>Henry, The inaccuracies stem from : (1) The shape functions are rational functions (derivatives will be more oscillatory) and hence Gauss quadrature can only do so much; and <strong>more importantly</strong>, (2) The intersection of basis function supports (region over which they are non-zero) do not coincide with the `background cell' that is used for integration. This is pertinent since to form the stiffness matrix one has to compute terms like \int \phi_i,x \phi_j,x dx in one dimension. The second item is particularly damaging . . . think of integrating (using Gauss quadrature) a continuous function that is zero from (-1,-0.1 ) and is a polynomial from (-0.1,1). If you use the `cells' : (-1,0) and (0,1) to integrate this function, Gauss quadrature will perform poorly. Of course, if you choose (-1,-0.1) and (-0.1,1) as your cells you can recover the exact answer. Here's a <a href="http://dolbow.cee.duke.edu/int.ps.gz" target="_blank">link to John's paper</a> (gzipped ps) that would provide more background, examples, and a clearer picture.</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Wed, 11 Apr 2007 06:55:50 +0000N. Sukumarcomment 2253 at https://imechanica.orgbackground grid for integral evaluation
https://imechanica.org/comment/2249#comment-2249
<a id="comment-2249"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>Dear Sukumar,</p>
<p>Can you elaborate on the first direction for the evaluation of the integral, which uses background cells?</p>
<p>Why this choice leads to inaccuracies in the patch test?</p>
<p>Can you suggest some references for further reading? Thanks a lot.</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Wed, 11 Apr 2007 01:50:11 +0000Henry Tancomment 2249 at https://imechanica.orgI would like to comment on
https://imechanica.org/comment/1619#comment-1619
<a id="comment-1619"></a>
<p><em>In reply to <a href="https://imechanica.org/comment/1049#comment-1049">Some tricky issues in meshfree method</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p class="MsoNormal"><font face="Times New Roman" size="3">I would like to comment on your use of “local form of Cosserat Balance laws “for avoiding Numerical Integration. I think this is only valid when the kinematic quantities are the functions of time only. For Cosserat surface and Cosserat rod approximations, kinematic quantities are functions of space variables also, so any weak projections needs Numerical integration, and one cannot avoid it. The only way to avoid it is to go for Collocation scheme. </font></p>
<p><font face="Times New Roman" size="3"> </font><br /></p><p class="MsoNormal"><font face="Times New Roman" size="3">Recently theory of Cosserat Points have been explored for nonlinear elasticity, where kinematic quantities are functions of time only, so that the resulting balance equations are nonlinear ODE’s instead of PDE’s. But the central difficulty lies in the unknown form of inhomogeneous strain energy functions, which cannot be obtained trivially for any class of materials. For homogeneous Cosserat Points, where deformation gradient is constant, the theory is well established, avoids the cost of Numerical Integration, which is even true for any linear finite elements. So, I think the issue of nodal integration can only be avoided if one uses the <strong>strong form</strong> of solutions. <span> </span>Please correct me if I am wrong in any of the above statements.</font></p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Sun, 11 Mar 2007 20:03:09 +0000B.Banerjeecomment 1619 at https://imechanica.orgpatch test for mesh free methods
https://imechanica.org/comment/1568#comment-1568
<a id="comment-1568"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>Where can i find useful information on patch test for Mesh free method.</p>
<p>thanks </p>
<p>raj</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Fri, 09 Mar 2007 15:50:28 +0000Amirtham Rajagopalcomment 1568 at https://imechanica.orgTricky issues (meshfree)
https://imechanica.org/comment/1135#comment-1135
<a id="comment-1135"></a>
<p><em>In reply to <a href="https://imechanica.org/comment/1049#comment-1049">Some tricky issues in meshfree method</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>All your points are valid; a few more supplements to what John has mentioned. Yes, issues such as numerical integration, `choice of support sizes' exist. Hence the attempt by many to move to stable nodal integration methods (at least for large deformation). A recent paper in ijnme (early view) provides a nice background on the subject, points our the issues, and proposes a solution. Yes, the isogeometric concept is appealing (given its ties to b-splines/nurbs) from the geometry-approximation perspective. If one wants freedom from a `mesh' one has to pay the price elsewhere . . . or at least be willing to do so. The link to the paper I mentioned is <a href="http://www3.interscience.wiley.com/cgi-bin/abstract/112775672/ABSTRACT" target="_blank" title="ijnme paper">here</a></p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Mon, 12 Feb 2007 05:56:13 +0000N. Sukumarcomment 1135 at https://imechanica.orgon the "tricky" meshfree issues
https://imechanica.org/comment/1096#comment-1096
<a id="comment-1096"></a>
<p><em>In reply to <a href="https://imechanica.org/comment/1049#comment-1049">Some tricky issues in meshfree method</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>I'll try to comment on most of your questions here.</p>
<p><em><font face="Times New Roman" size="3">1. In the refinement process one needs to modify the background mesh. Is this not numerically intensive?</font></em></p>
<p>It's not necessarily the case that the background mesh needs to be modified with refinement. Many methods do use a *fixed* background mesh. Further, some methods employ a variation on nodal integration, and do not rely on a background mesh at all. Finally, the modification of the mesh, when it is used, need not be numerically intensive<font face="Times New Roman" size="3">.</font></p>
<p><font face="Times New Roman" size="3"><em>2. Since Meshfree shape functions are rational we need higher order quadrature rule for numerical integration. This again increases the computation time.</em></font></p>
<p>Not all meshfree shape functions are rational functions. Sometimes people do use high-order Gauss quadrature, and that certainly can be expensive. However, I would contend that this is not a particularly smart way to handle things. </p>
<p><em><font face="Times New Roman" size="3">3. Support size of window function is one of the very tricky issue in Meshfree method. User requires some numerical experiment to fix the optimum support size.</font> </em></p>
<p>This is a valid point. However, some of the "better" meshfree methods do not have this problem. For example, with NEM, there is no support size question. </p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Fri, 09 Feb 2007 21:20:20 +0000John E. Dolbowcomment 1096 at https://imechanica.orgSome tricky issues in meshfree method
https://imechanica.org/comment/1049#comment-1049
<a id="comment-1049"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p class="MsoNormal"><font face="Times New Roman" size="3">I would like to add few comments. Firstly as we know that all difficulties present in FEM are consequence of use of elements (mesh) which are essentially the building blocks of FEM. Meshfree method is introduced as an alternative scheme to overcome these difficulties. However, even in Meshfree methods based on weak projection of the governing PDE, one requires a background mesh. Consequently,</font></p>
<p><font face="Times New Roman" size="3"> </font>
</p><ol><li class="MsoNormal"><font face="Times New Roman" size="3">In the refinement process one needs to modify the background mesh. Is this not numerically intensive?</font></li>
<li class="MsoNormal"><font face="Times New Roman" size="3">Since Meshfree shape functions are rational we need higher order quadrature rule for numerical integration. This again increases the computation time. </font></li>
<li class="MsoNormal"><font face="Times New Roman" size="3">Support size of window function is one of the very tricky issue in Meshfree method. User requires some numerical experiment to fix the optimum support size.</font></li>
</ol><p><font face="Times New Roman" size="3"> </font><font face="Times New Roman" size="3">Now the question is “is it possible to bypass the above mentioned difficulties?”</font><font face="Times New Roman" size="3"> </font><br /></p><p class="MsoNormal"><font face="Times New Roman" size="3"><strong>Issue of integration:</strong> </font></p>
<p class="MsoNormal"><font face="Times New Roman" size="3">One may go for collocation technique. But this may lead to stability problem. I think one solution could be the use of Cosserat continuum theory where balance laws are formulated based on some deformable directors. Using the local form of balance laws one may easily avoid the burden of numerical integration. Moreover it is already numerically demonstrated that Cosserat theory does not suffer any stability problem.</font></p>
<p><font face="Times New Roman" size="3"> </font><br /></p><p class="MsoNormal"><font face="Times New Roman" size="3"><strong>Issue of support size</strong></font></p>
<p class="MsoNormal"><font face="Times New Roman" size="3">One may use a window function that having automatically adjustable support size depending on the order of consistency. One immediate choice could be the use of non-uniform rational B-splines (NURBS). <span> </span>You may see this paper “A NURBS based error reproducing kernel method”, Appeared online in computational mechanics.</font></p>
<p><font face="Times New Roman" size="3"> </font><br /></p><p class="MsoNormal"><font face="Times New Roman" size="3"><strong>Connection between geometry and approximation:</strong></font></p>
<p class="MsoNormal"><font face="Times New Roman" size="3">Sukumar, as you have told that there must be a relation between geometry and approximation. In this regard the iso-geomeric analysis concept proposed by Hughes et all is interesting (CMAME, vol 194, issue 39-41, pp.4135-4195).</font></p>
<p><font face="Times New Roman" size="3"> </font><br /></p><p class="MsoNormal"><font face="Times New Roman" size="3">I welcome any further comments on this.</font></p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Wed, 07 Feb 2007 20:15:29 +0000abshawcomment 1049 at https://imechanica.orgA truly meshfree method
https://imechanica.org/comment/623#comment-623
<a id="comment-623"></a>
<p><em>In reply to <a href="https://imechanica.org/node/490">2. Is a mesh required in meshfree methods?</a></em></p>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even"><p>Dear Sukumar,</p>
<p>Thank you very much for your excellent posts both here and on meshfreemethods blogspot. Below are my opinions.</p>
<p>- I think local Voronoi mesh used in nodal integration is also a mesh. However, to mesh and adjust in a local domain are not time-consuming tasks such as in FEM. Moreover, with strain smoothing technique, this method can give very good results.</p>
<p>- If one uses Galerkin methods and does not want to deal with meshes in numerical integration, he can use moving integration subdomains. In this method, a set of patches <strong>Ω</strong>k covers the entire problem domain is used with its corresponding distribution function <strong>Ψ</strong>k . </p>
<p>The patches can overlap each other or not (but there is no gap). A simple geometry for <strong>Ω</strong>k often used is quadrature (in 2D). If the centers of these quadratures coincide with given nodes, then <strong>Ω</strong>k(moving quadratures at node <strong>X</strong>k) must be covered by the support domain of node <strong>X</strong>k. They also can be the same. </p>
<p>The <em>Shepard </em>function is one of the distribution functions, which partition the unity.</p>
<p>Using these techniques leads to integrate on subdomains (moving quadratures) in stead of the entire domain when computing each nodal stiffness/force matrix. With linear problems, just one Gauss point is needed for each subdomain, and it is very interesting that it is the given node.</p>
<p> That is one of truly meshfree methods for Galerkin weak form. No mesh is needed. </p>
<p>Quoc-Duan</p>
</div></div></div><ul class="links inline"><li class="comment_forbidden first last"><span><a href="/user/login?destination=node/490%23comment-form">Log in</a> or <a href="/user/register?destination=node/490%23comment-form">register</a> to post comments</span></li>
</ul>Tue, 26 Dec 2006 16:45:42 +0000Nguyen Quoc Duancomment 623 at https://imechanica.org