Hello,
I built a simple dipole antenna to study the feature of the Livelink for MATLAB. The dipole is built as a 2D axisymmetric structure. Both eigenvalue solver and eigenfrequency solver are adopted for computation and matched eigenvalues are achieved between the two solvers (f=-lambda/(2*pi*i)). After the simulation in COMSOL, I exported the corresponding M-file and ran it in the COMSOL 4.3 with MATLAB. By implementing the following command, I tried to extract the eliminated system matrix:
MK = mphmatrix(model, 'sol1', 'Out', {'Kc','Dc'});
The problem is that every time I ran the M-file exported from COMSOL, I got different MK.Kc and MK.Dc while the code is the same. Is there any one have similar problem and how do you deal with it?
Furthermore, I also tried to get the eigenvalue of the structure by the following MATLAB command:
[ eigvec , omega]= eigs (MK.Kc, MK.Dc, 10,-8e7*2*pi*1i) ;
The eigenvalue calculated from above is quite different from the results computed by COMSOL directly. Any one got an idea about the issue? To have a better description, I attached the two m-files I mentioned for your reference. Thanks for your help in advance.
I built a simple dipole antenna to study the feature of the Livelink for MATLAB. The dipole is built as a 2D axisymmetric structure. Both eigenvalue solver and eigenfrequency solver are adopted for computation and matched eigenvalues are achieved between the two solvers (f=-lambda/(2*pi*i)). After the simulation in COMSOL, I exported the corresponding M-file and ran it in the COMSOL 4.3 with MATLAB. By implementing the following command, I tried to extract the eliminated system matrix:
MK = mphmatrix(model, 'sol1', 'Out', {'Kc','Dc'});
The problem is that every time I ran the M-file exported from COMSOL, I got different MK.Kc and MK.Dc while the code is the same. Is there any one have similar problem and how do you deal with it?
Furthermore, I also tried to get the eigenvalue of the structure by the following MATLAB command:
[ eigvec , omega]= eigs (MK.Kc, MK.Dc, 10,-8e7*2*pi*1i) ;
The eigenvalue calculated from above is quite different from the results computed by COMSOL directly. Any one got an idea about the issue? To have a better description, I attached the two m-files I mentioned for your reference. Thanks for your help in advance.