In an earlier post on simulation based product design we saw that in many cases a large portion of effort early on in the design cycle is usually spent on determining the best practices to simulate a physical event. We can speed up such simulations by the discrete variable support in LS-OPT version 3.1. An example of the LSOPT command file in which two parameters in LS-DYNA, SOFT and SSTHK, are defined as discrete variables, is included here. The command file can be viewed using LSOPTUI – a graphical user interface for LSOPT.
"Discrete Optimization Example"
Author "Suri Bala"
$ Created on Mon Apr 30 10:55:18 2007
solvers 1
responses 2
$
$ NO HISTORIES ARE DEFINED
$
$
$ DESIGN VARIABLES
$
variables 2
Variable 'soft' 0
Variable 'soft' discrete {0 1 2 }
Variable 'ssthk' 0
Variable 'ssthk' discrete {0 1 }
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$ OPTIMIZATION METHOD
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
Optimization Method SRSM
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$ SOLVER "discrete"
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
$ DEFINITION OF SOLVER "discrete"
$
solver own 'discrete'
solver command "dyna.exe -i"
solver input file "dyna_input.k"
$ ------ Pre-processor --------
$ NO PREPROCESSOR SPECIFIED
$ ------ Metamodeling ---------
solver order linear
solver experiment design dopt
$ ------ Job information ------
$
$ RESPONSES FOR SOLVER "discrete"
$
response 'sliding_energy' 1 0 "BinoutResponse -res_type GLStat -cmp sliding_interface_energy -select TIME "
response 'internal_energy' 1 0 "BinoutResponse -res_type GLStat -cmp internal_energy -select TIME "
$
$ OBJECTIVE FUNCTIONS
$
objectives 1
maximize
objective 'internal_energy' 1
$
$ CONSTRAINT DEFINITIONS
$
constraints 1
constraint 'sliding_energy'
lower bound constraint 'sliding_energy' 0
$
$ JOB INFO
$
iterate param design 0.01
iterate param objective 0.01
iterate param stoppingtype or
iterate 1
STOP
