Optimization using modal analysis failing
 

Hello all!

I am a student currently working on an optimization problem for my thesis in which I have to optimize a composite material rectangular plate regarding eigenfrequencies (program should calculate the best fiber orientation scheme possible from a number of choises using DMO method).

I have written all the code required for this scheme to work but I am however having a problem with a specific step. I need to calculate the product of the modes of vibration with the global stiffness matrix (Mode'*[K]*Mode). However the result of this step is coming out wrong and resulting in the Matlab optimizer not finding the steepest descent direction.

I am using the hbmat command to extract the stiffness matrix from Ansys:
aux2
file,file,full
hbmat,K,txt,,ascii,stiff,no,yes
fini

Afterward since the modal analysis is normalized on the mass I am extracting the nodal displacements and rotations to Matlab and then creating a mode of vibration vector with them by using the mapping information contained in K.mapping so that the DOF order is correctly reflected.

The displacements and rotations are extracted with:
/OUTPUT,Ux,txt
PRNSOL,U,X
/out
etc...

I suspect there is a mistake somewhere on this procedure but cannot find it. Does anyone have a clue about what may be wrong?

I assumed the mapping file's field "Matrix Eqn" refers to the matrix entry (line column) and the "Node" and "DoF" fields refer to the corresponding nodal information.

Sorry for the long post and thanks in advance for the attention!