3. Principle of operation#

Depending on its input arguments, CALC_MISS produces a concept whose type varies or does not produce a concept.

  • If TYPE_RESU is set to “FICHIER” or “FICHIER_TEMPS”, no concept is produced. Only the execution of Miss3D is started. The results (ground impedance and seismic forces) are then written into the files identified by the logic units such as UNITE_RESU_IMPE, UNITE_RESU_FORC, UNITE_RESU_MASS, UNITE_RESU_RIGI or UNITE_RESU_AMOR. There is no post-processing of the results from Miss3D.

  • If TYPE_RESU = “CHARGE”, a mechanical load is produced in the form of a nodal force.

  • If TYPE_RESU = “TABLE_CONTROL”, the Miss3D calculation is the same as for FICHIER. A table is produced containing a one-time post-processing of the Miss3D results.

  • Otherwise (TYPE_RESU is equal to “HARM_GENE”, “”, “TRAN_GENE” or “TABLE”), Miss3D is executed only if the logical units UNITE_RESU_IMPE, UNITE_RESU_FORC are not specified. Otherwise, we use the files provided. Post-processing is then performed and the requested concept returned to the user.

MISS3D calculations on big models (or with a lot of frequencies) can be time consuming and memory intensive. Fortunately, these can be accelerated by activating one or two levels of parallelism. The value TYPE_RESU =” FICHIER_TEMPS “allows you to activate two, the other values, only one. For more information you can consult the documentation [U2.06.07] and [U2.08.06].

CALC_MISS produces a tran_gene or harm_gene concept, including acceleration and absolute displacement fields. In this case, the accelerations are directly used to calculate the floor spectrum; conversely, the deduction of ground displacements is mandatory (to be read with the command LIRE_FONCTION [U4.32.02] from a specific file given by its logical unit) to obtain the relative displacements.

_images/100000000000042400000664EF0394177EEB374C.png

When running Miss3D, if the MACR_ELEM_DYNA keyword is entered, it is used. Otherwise, it’s created by CALC_MISS using the BASE_MODALE, MATR_RIGI, and MATR_MASS operands.

Note

In the case FICHIER_TEMPS , we call Miss3D for each calculation frequency. These calls can be made in parallel. To do this, simply run parallel version MPI of Code_Aster and ask for several processors (no additional keyword required) .