Hi, I have a question to propose.
As the eigenvectors are determined?
What is the rule of normalization of the eigenvectors?
Can anyone give me some help?
Thanks so much.
eigenvector
Moderators: silvia, selimgunay, Moderators
Re: eigenvector
Depending on the solver you use to solve the eigenvalue problem the normalization will be different. It depends on the solver package, it is not under our control.
-
- Posts: 23
- Joined: Tue Nov 16, 2010 6:14 am
- Location: Sapienza
Re: eigenvector
Thanks for your reply.
For example, if the command object in the analysis are:
constraints Lagrange ;
numberer RCM;
system BandGeneral ;
test NormDispIncr 1.0e-8 6 ;
algorithm Newton;
integrator LoadControl 10;
analysis Static;
eigen frequency 3
analyze 0.1;
For the following example which is the type of normalization used for the eigenvectors?
Thanks.
For example, if the command object in the analysis are:
constraints Lagrange ;
numberer RCM;
system BandGeneral ;
test NormDispIncr 1.0e-8 6 ;
algorithm Newton;
integrator LoadControl 10;
analysis Static;
eigen frequency 3
analyze 0.1;
For the following example which is the type of normalization used for the eigenvectors?
Thanks.
Re: eigenvector
Eigen analysis has its own solver that you specify when you call eigen command. The example you posted does not define the solver for eigen analysis, it defines analysis objects for static analysis. In your example the following line:
eigen frequency 3
is incorrect.
Please read the command manual to see what arguments you need to specify to define eigen analysis:
http://opensees.berkeley.edu/wiki/index ... en_Command
If you are really interested to see how each of the solvers normalizes the eigenvectors look at the source code:
http://opensees.berkeley.edu/WebSVN/lis ... d5631c47c4
eigen frequency 3
is incorrect.
Please read the command manual to see what arguments you need to specify to define eigen analysis:
http://opensees.berkeley.edu/wiki/index ... en_Command
If you are really interested to see how each of the solvers normalizes the eigenvectors look at the source code:
http://opensees.berkeley.edu/WebSVN/lis ... d5631c47c4
-
- Posts: 23
- Joined: Tue Nov 16, 2010 6:14 am
- Location: Sapienza
Re: eigenvector
Thanks so much.