I am currently implementing some meshfree methods in Matlab. I have already implemented the EFG method. Now I'm trying to implement the MLPG mixed collocation method, and here I run into a problem.

I do not know how to apply the essential boundary conditions. In the EFG method I have used lagrangian multipiers to apply the essential boundary conditions, however with the MLP mixed collocation method you are not working with the weak form (or the local weak form in the normal MLPG method), but with the momentum equations. Thus, I don't think I can use the lagrangian multipliers for the essential boundary conditions. Does anyone have an idea, or is anyone familiar with implementing the MLPG mixed collocation method?

Hi All,

A minisymposium on Meshfree and Particle Methods will be organized for the International
Conference on Computational Methods (ICCM) 2010. ICCM 2010 will be organized and hosted by
College of Mechanical & Vehicle Engineering, Hunan University during November 19th-21st, 2010 in
Zhangjiajie, Hunan Province, China. Please check the ICCM 2010 website
http://dmvb.hnu.cn/ICCM2010 for more information about the conference and the venue.

Hi all,

The beginers to the EFG method may get quick start inside the method through the simple EFG code for bars and beams here !

Regards,

 Canh Le

Hello,

I wish to ask where to find more references about application of constrained moving least squares method for imposing displacement bounday conditions in meshless methods. The most important advantage of CMLS over MLS is satisfying Kronecker delta function property, Hence such as finite element methods we can impose the essential boundary conditions in meshless method using CMLS approach.

Thank you,

Jafar Amani

Attached is a tar archive for a Fortran 90 library to compute maximum-entropy basis functions.  I have used the G95 compiler. The manual in PDF is also attached and a html version of the same is also available, which provide details on how to install the code and its capabilities. This library was tested by Mike Puso last year, who interfaced it to DYNA3D and NIKE3D codes.  Note that I have added a .doc extension to the tar file (can not upload files with .tar extension).

I just wanted to know if i can consider one part of a FE model as a meshless part and form the global stiffness matrix just by assembling the meshfree stiffness matrix corresponding to the meshfree zone and the FE stiffness matrix of the rest. And then apply the boundary conditions to the model and solve. I would use RKPM to generate the meshfree stiffness matrix. 

for example mesh a square plate into 4  elements , suppose ele 1,2,3 are e finite elements and  the rest of the domain is discretized using nodes. such that there are 4 nodes on the four corners of the forth piece. and many other nodes in the interior.

An Australian Research Council funded PhD Scholarship is available in the Department of Civil Engineering at Monash University in Australia in the area of computational mechanics. The objective of this project is to develop a multi-scale bifurcation-based decohesion model within the framework of the Material Point Method (MPM), one of the meshfree methods, for simulating glass fragmentation under blast loading. The proposed multi-scale decohesion model will be calibrated by combining molecular dynamics and continuum mechanics approaches, and the simulation results will be verified by available experimental data.

Dear iMechanica colleagues,

I would like to announce that Matlab routines implementing the maximum-entropy approximation schemes presented in

Marino Arroyo and Michael Ortiz, “Local maximum-entropy approximation schemes: a seamless bridge between finite elements and meshfree methods”, International Journal for Numerical Methods in Engineering, 65:2167–2202 (2006).

can be downloaded from