Optimization Preprocessing Modules

Within this section, some preprocessing-related topics are discussed.

This page discusses:

FEM_INCLUDE

This module parses the FEM_INCLUDE command of a parameter file and generates a list of all files that are needed for the optimization. These files are copied from the input directory into the working directory of the current optimization job.

FEM_2_ONF

This module converts all solver-specific model files into ONF format for later processing.

TOSCA_PREP

In TOSCA_PREP, the analysis model and the commands used to define the optimization problem are loaded to create the optimization model. The analysis model exists as a solver input model and is converted before loading into ONF format using the FEM2ONF module.

The optimization problem is defined by either using Tosca Structure.pre or an editor. After the data has been loaded, an internal check run is carried out to ensure that the optimization model is complete. If the check run is completed successfully, an optimization database is created which is always referred to during the optimization. When the optimization preprocessor starts, the control program loads the file $tosca/command.ini automatically. This file contains the presettings for the optimization.

Dynamic Memory Management:

Tosca Structure offers a dynamic memory management, that is, the necessary memory requirement depends on the size of the model and is read from the configuration file when the program is started. The required disk space depends on the size of the finite element model, or more accurately, on the number of elements or nodes in the analysis model. In case of larger numbers of elements or nodes, an automatic reallocation of required memory is performed. The applied default settings for the memory allocation should be sufficient in order to avoid this reallocation for most models.