Importing and Exporting Optimization Configuration Files

You can import and export an Optimization configuration file. The subsequent sections show an example and the specific options for each available technique. To allow for better formatting in the file examples, the “*” and “#” symbols are used in various places to denote substitutions for defined values.

This page discusses:

Optimization Configuration File Example

Technique: NLPQLP
Technique Options:
Max Iterations: 10
Termination Accuracy: 1.0E-6
Rel Step Size: 0.0010
Min Abs Step Size: 1.0E-4
Use Central Differences: false
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Save Technique Log: false
Execution Options:
PARAM_TERMINATION_ALL: false
parallel: false
PARAM_TERMINATION_SKIP_RUNS: 0
Re-execute optimum point: true
TERMINATE_BY_FEASIBILITY: -1
PARAM_TERMINATION_CONSECUTIVE_RUNS: 1
Restore best design: true
TERMINATE_BY_THE_CLOCK_TIME: 0
TERMINATE_BY_PARAMETERS: false
ELAPSED_TIME_UNITS: 0
Use automatic scaling: false
TERMINATE_BY_ELAPSED_TIME: -1
execute subflow once: false
import at runtime: false
TERMINATE_BY_RUNS: -1
Design Parameters:
Design Variables:
Name "Lower Bound" "Upper Bound" "Gene Size" "Scale Factor"
"*" 0.9999 20.0 16 1.0
"*" 0.04999 1.0 16 1.0
"***" 0.09999 1.0 16 1.0 
"*" – NumberOfCoils
"**" – WireDiameter
"***" – CoilDiameter
Output Constraints:
Name "Upper Bound" "Scale Factor" "Weight Factor"
ShearStress 0.0 1.0 1.0
Size 0.0 1.0 1.0
Deflection 0.0 1.0 1.0
SurgeFrequency 0.0 1.0 1.0 
Objectives:
Name Direction "Scale Factor" "Weight Factor"
Weight minimize 1.0 1.0

Adaptive DOE Technique Options

Technique: Adaptive DOE
Technique Options:
Number of Adaptive Iterations: 10
Number of Points per Iterations: 20
Significant Step Size: 0.01
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Use fixed random seed: false
Random Seed Value: -1

Adaptive Simulated Annealing Technique Options

Technique: ASA
Technique Options:
Max Number of Generated Designs: 10000
Number of Designs for Convergence Check: 5
Convergence Epsilon: 1.0E-8
Relative Rate of Parameter Annealing: 1.0
Relative Rate of Cost Annealing: 1.0
Relative Rate of Parameter Quenching: 1.0
Relative Rate of Cost Quenching: 1.0
Max Number of Failed Designs: 5
Init Param Temperature: 1.0
Reanneal Parameters: true
Reanneal Cost Function: true
Num of Gener Designs Before Reannealing: 1000
Num of Accepted Designs Before Reannealing: 100
Min Ratio of Accepted Designs for Reannealing: 1.0E-6
Rel Gradient Step for Param Reannealing: 0.0010
Penalty Base: 0.0
Penalty Multiplier: 1000.0
Penalty Exponent: 2
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Use fixed random seed: false
Random Seed Value: -1

Archive-Based Micro Genetic Algorithm Technique Options

Technique: AMGA
Technique Options:
Initial Size: 40
Population Size: 40
Number of Function Evaluations: 500
Archive Size Limit: 500
Pareto Size Limit: 100
Crossover Probability: 0.9
Mutation Probability: 0.5
Use optimal mutation probability: true
Crossover Distribution Index: 10.0
Mutation Distribution Index: 20.0
Use fixed random seed: false
Random seed value: -1
Initialization Mode: Random
Initialization Filename: null
Diversity Option: Crowding

DownhillSimplex Technique Options

Technique: DownhillSimplex
Technique Options:
Initial Simplex Size: 0.1
Max Iterations: 40

Evolutionary Optimization Algorithm Technique Options

Technique: Evol Technique
Technique Options:
Max Evaluations: 100
Convergence Tolerance: 0.01
Minimum Discrete Step: 0.02
Consecutive Variable Search: false
Parallel Batch Size: 10
Penalty Base: 0.0
Penalty Multiplier: 1000.0
Penalty Exponent: 2
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Use Fixed Random Seed: true
Random Seed Value: 12345

Generalized Reduced Gradient Technique Options

Technique: LSGRG
Technique Options:
Max Iterations: 10
Convergence Epsilon: 0.0010
Rel Step Size: 0.0010
Convergence Iterations: 3
Binding Constraint Epsilon: 1.0E-4
Phase 1 Objective Ratio: 1.0
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Save Technique Log: false

Hooke-Jeeves Direct Search Method Technique Options

Technique: Hooke-Jeeves
Technique Options:
Max Iterations: 10
Max Evaluations: 100
Relative Step Size: 0.5
Step Size Reduction Factor: 0.5
Termination Step Size: 1.0E-6
Penalty Base: 0.0
Penalty Multiplier: 1000.0
Penalty Exponent: 2
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Save Technique Log: false

Multi-Island Genetic Algorithm Technique Options

Technique: Multi-Island GA
Technique Options:
Sub-Population Size: 10
Number of Islands: 2
Number of Generations: 4
Rate of Crossover: 1.0
Rate of Mutation: 0.01
Rate of Migration: 0.5
Interval of Migration: 2
Elite Size: 1
Rel Tournament Size: 0.5
Penalty Base: 0.0
Penalty Multiplier: 1000.0
Penalty Exponent: 2
Max Failed Runs: 5
Failed Run Penalty Value: 1000000.0
Failed Run Objective Value: 1000000.0
Default Variable Bound (Abs Val): 1000.0
Use fixed random seed: false
Random Seed Value: -1

Multifunction Optimization System Tool Technique Options

Technique: MOST
Technique Options:
Max Iterations: 10
Convergence Tolerance: 1.0E-4
Rel Gradient Step: 1.0E-4
Min Absolute Step: 1.0E-4
Max Confirmation Runs: 5
Min Obj Change: 1.0E-10
Constraint Tolerance: 1.0E-4
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30

Multi-Objective Particle Swarm Technique Options

Technique: Multi-Objective Particle Swarm
Technique Options:
Maximum Iterations: 50
Number of Particles: 10
Inertia: 0.9
Global Increment: 0.9
Particle Increment: 0.9
MaxMin Velocity: 0.1
Use Fixed Random Seed: true
Random Seed Value: 12345

Non-dominated Sorting Genetic Algorithm Technique Options

Technique: NSGA-II
Technique Options:
Population Size (even value): 12
Number of Generations: 20
Crossover Probability: 0.9
Crossover Distribution Index: 10.0
Mutation Distribution Index: 20.0
Use fixed random seed: false
Random seed value: -1
Initialization Mode: Random
Initialization Filename: null

Pointer Automatic Optimizer Technique Options

Technique: Pointer
Technique Options:
Maximum allowable job time (hr): 1.0
Average analysis time (sec): 1.0
Topography type: nonlinear
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30
Save Technique Log: false
Use fixed random seed: false
Random Seed Value: -1

Python Optimizer Technique Options

Import/Export of technique configuration is not supported for Python Optimizer.

Sequential Quadratic Programming Technique Options

Technique: NLPQLP
Technique Options:
Max Iterations: 10
Termination Accuracy: 1.0E-6
Rel Step Size: 0.0010
Min Abs Step Size: 1.0E-4
Use Central Differences: false
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30

Stress Ratio Technique Options

Technique: Stress Ratio
Technique Options:
Max Iterations: 20
Power: 0.4
Constraint Critical Ratio: 1.0
Tolerance: 0.0
Max Failed Runs: 5
Failed Run Penalty Value: 1.0E30
Failed Run Objective Value: 1.0E30