Personal tools
 

History MSE: multi-case

In this example time history curves are used to calibrate directly experiment vs. simulation curves with two analysis cases.

 

Solution with LS-OPTui

Solution with LS-OPTui

Solvers

Solvers

Start LS-OPTui.

→ Open the command file com.mse_history generated in param_history.

Solvers Panel

 

  1. Select the Solvers panel.
  2. For Command specify the LS-DYNA executable ls971_R4_2 (This name can be different on your computer).
  3. For Input File browse the file foam2.k in your directory.
  4. Enter the Name of Analysis Case: Case2.
  5. Push the Add button to create a new analysis case.
 solvers_case22.png

Sampling

Sampling

Sampling

 

  1. Select the Sampling panel.
  2. Select the analysis case Case2 from the list.
  3. For POINT SELECTION select Duplicate.
  4. For Case choose Case1.
sampling2.png

Histories

Histories

Define Response Histories

 

Define z-displacement of node 288

  1. Select the Histories panel.
  2. From the possible histories select NODOUT.
  3. For analysis Case select Case2.
  4. Type in the node ID: 288.
  5. For Direction select Z Component.
  6. Type in Disp2 for the History Name.
  7. Push the Add button.

 

 

history_disp2.png

Define z-slave-force on interface 1

  1. From the possible histories select RCFORC.
  2. Type in for Interface ID: 1.
  3. For Component select Z slave force.
  4. Type in Force2 for the History Name.
  5. Push the Add button.
history_force2.png
Define crossplot for -Disp2 and Force2
  1. From the possible histories select Crossplot.
  2. For z(t) type in -Disp2 (negative Disp1).
  3. For F(t) type in Force2.
  4. Type in F2_vs_d2 as History Name.
  5. Push the Add button to create a new history.
history_f2-vs-d2.png
Define a history through a text file
  1. From the possible histories select File.
  2. Push the Browse button to find the file Test2.txt from your main directory.
  3. Type in FILE2 as History Name.
  4. Push the Add button to create a new history.
history_test2.png

 

Responses

Responses

Define the Responses

 

  1. Select the Responses panel.
  2. From the possible response types select: MeanSqErr.
  3. For Target curve select FILE2.
  4. For Computed curve select F2_vs_d2.
  5. Type in the label MSE2 for Response Name.
  6. Push the Add button.
response_mse2.png

Objective

Objective

Objectives

 

  1. Select the Objective panel.
  2. Select MSE from Response as the objective function.
  3. Leave the default 1.0 for Weight.
  4. Select MSE2 from Response as the objective function.
  5. Leave the default 1.0 for Weight.
objective_mse2.png

Run

Run

Run the Optimization

 

  1. Select the Run panel.
  2. For Number of iterations enter 3.
  3. Push the Run button to start the optimization.  (Save the project as com.mse_history_multi)
run2.png

Com-file

Com-file

The created command file may look like this:

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Command file "com.mse_history_multi"
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$ Generated using LS-OPT Version 4.1
$
"Optimization Problem"
$
$ Created on Fri Jan 21 17:10:51 2011
solvers 2
$
$ WARNING -- NO RESPONSES ARE DEFINED
$
histories 8
$
$ DESIGN VARIABLES
$
variables 2
 Variable 'YMod' 7.e5
  Lower bound variable 'YMod' 5.e5
  Upper bound variable 'YMod' 2.e6
 Variable 'Yield' 1500.
  Lower bound variable 'Yield' 500.
  Upper bound variable 'Yield' 2.e3

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$      OPTIMIZATION METHOD   
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
Optimization Method SRSM

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$      SOLVER "Case1"
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
$ DEFINITION OF SOLVER "Case1"
$
 solver dyna960 'Case1'
  solver command "ls971_R4_2"
  solver input file "foam1.k"
  solver check output on 
  solver compress d3plot off 
$ ------ Pre-processor --------
$   NO PREPROCESSOR SPECIFIED
$ ------ Post-processor --------
$   NO POSTPROCESSOR SPECIFIED
$ ------ Metamodeling ---------
  solver order linear
  solver experiment design dopt
  solver approximate history linear
$ ------ Job information ------
  solver concurrent jobs 1
$
$ WARNING - NO RESPONSES DEFINED FOR SOLVER "Case1"
$
$
$ HISTORIES FOR SOLVER "Case1"
$
 history 'Disp1' "BinoutHistory -res_type nodout  -cmp z_displacement -id 296"
 history 'Force1' "BinoutHistory -res_type RCForc -cmp z_force  -id 1  -side SLAVE"
$
$ HISTORY EXPRESSIONS FOR SOLVER "Case1"
$
 history 'F_vs_d' expression {Crossplot("-Disp1","Force1")}

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$      SOLVER "Case2"
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
$ DEFINITION OF SOLVER "Case2"
$
 solver dyna960 'Case2'
  solver command "ls971_R4_2"
  solver input file "foam2.k"
  solver check output on 
  solver compress d3plot off 
$ ------ Pre-processor --------
$   NO PREPROCESSOR SPECIFIED
$ ------ Post-processor --------
$   NO POSTPROCESSOR SPECIFIED
$ ------ Metamodeling ---------
  solver order linear
   solver experiment duplicate 'Case1'
  solver approximate history linear
$ ------ Job information ------
  solver concurrent jobs 1
$
$ WARNING - NO RESPONSES DEFINED FOR SOLVER "Case2"
$
$
$ HISTORIES FOR SOLVER "Case2"
$
 history 'Disp2' "BinoutHistory -res_type nodout  -cmp z_displacement -id 288"
 history 'Force2' "BinoutHistory -res_type RCForc -cmp z_force  -id 1  -side SLAVE"
$
$ HISTORY EXPRESSIONS FOR SOLVER "Case2"
$
 history 'F2_vs_d2' expression {Crossplot("-Disp2","Force2")}

$
$ HISTORIES FROM FILES
$
 history 'FILE1' file "Test1.txt"
 history 'FILE2' file "Test2.txt"
composites 2
$
$ COMPOSITE EXPRESSIONS
$
 composite 'MSE' {MeanSqErr(FILE1,F_vs_d)}
 composite 'MSE2' {MeanSqErr(FILE2,F2_vs_d2)}
$
$ OBJECTIVE FUNCTIONS
$
 objectives 2
 objective 'MSE' 1
 objective 'MSE2' 1
$
$ THERE ARE NO CONSTRAINTS!!!
$
 constraints 0
$
$ PARAMETERS FOR METAMODEL OPTIMIZATION
$
 Metamodel Optimization Strategy DOMAINREDUCTION
$
  iterate param design 0.01
  iterate param objective 0.01
  iterate param stoppingtype and
$
$ OPTIMIZATION ALGORITHM
$
 Optimization Algorithm hybrid simulated annealing
  Use GSA
$
$ JOB INFO
$
 iterate 3
STOP

Results

Results

Histories

Histories

Start the LS-OPT Viewer by selecting the View tab and

 

  1. Select under Simulations the item Histories.
  2. From the current window open a new plot.
  3. Select Placement to put the new plot below the old one.
  4. Select the type of plot Histories under Simulations.

new_plot1.png

menubar.png

new_plot2.png

Set up the left panel for Case1

 

  1. Select the last iteration.
  2. Set time as x-coordinate.
  3. Choose history response F_vs_d as y-coordinate.
  4. Add FILE1 to y-coordinate, so that we can compare its plot with F_vs_d 's.
  5. For the third coordinate, whose quantity is displayed by means of colors, we choose variable YMod. (Or other interesting entities)

 

Set up the right panel for Case2

 

  1. Select the last iteration.
  2. Set time as x-coordinate.
  3. Choose history response F_vs_d2 as y-coordinate.
  4. Add FILE2 to y-coordinate, so that we can compare its plot with F_vs_d2 's.
  5. For the third coordinate (the color coordinate), we choose variable YMod. (Or other interesting entities)

 

history_panel.png

 → It's very convenient to compare them, if we place the plots from two cases in one diagram.

history_plots.png

 

Confidence Intervals

Confidence Intervals

The Confidence Intervals can be found in the lsopt_report file.

  1. From the main menu bar select View → Summary Report.

 

The Confidence Intervals are located at the end of the file.

view_summary2.png

Alternatively, we can directly open the lsopt_report file from the main directory with an editor and scroll to the end to find the Confidence Intervals of YMod and Yield.

→ YMod is rather insignificant and Yield is significant.

conf_interval2.png

Download

Download

The complete data set (input and command files) is available for download:

For Linux

For Windows