soga

Single-objective Genetic Algorithm (a.k.a Evolutionary Algorithm)

Topics

package_jega, global_optimization_methods

Specification

• Alias: None

• Arguments: None

Child Keywords:

Required/Optional	Description of Group	Dakota Keyword	Dakota Keyword Description
Optional		fitness_type	Select the fitness type for JEGA methods
Optional		replacement_type	Select a replacement type for JEGA methods
Optional		convergence_type	Select the convergence type for JEGA methods
Optional		max_iterations	Number of iterations allowed for optimizers and adaptive UQ methods
Optional		max_function_evaluations	Number of function evaluations allowed for optimizers
Optional		scaling	Turn on scaling for variables, responses, and constraints
Optional		population_size	Set the initial population size in JEGA methods
Optional		log_file	Specify the name of a log file
Optional		print_each_pop	Print every population to a population file
Optional		initialization_type	Specify how to initialize the population
Optional		crossover_type	Select a crossover type for JEGA methods
Optional		mutation_type	Select a mutation type for JEGA methods
Optional		seed	Seed of the random number generator
Optional		convergence_tolerance	Stopping criterion based on objective function or statistics convergence
Optional		model_pointer	Identifier for model block to be used by a method

Description

soga stands for Single-objective Genetic Algorithm, which is a global optimization method that supports general constraints and a mixture of real and discrete variables. soga is part of the JEGA library.

Constraints soga can utilize linear constraints.

Configuration

The genetic algorithm configurations are:

• fitness

• replacement

• convergence

• initialization

• crossover

• mutation

• population size

The pool of potential members is the current population and the current set of offspring. Choice of fitness assessors is strongly related to the type of replacement algorithm being used and can have a profound effect on the solutions selected for the next generation.

Stopping Criteria

The soga method respects the max_iterations and max_function_evaluations method independent controls to provide integer limits for the maximum number of generations and function evaluations, respectively.

The algorithm also stops when convergence is reached. This involves repeated assessment of the algorithm’s progress in solving the problem, until some criterion is met.

Expected Outputs The soga method respects the output method independent control to vary the amount of information presented to the user during execution.

The final results are written to the Dakota tabular output. Additional information is also available - see the log_file and print_each_pop keywords.

Expected HDF5 Output

If Dakota was built with HDF5 support and run with the hdf5 keyword, this method writes the following results to HDF5:

• Best Parameters

• Best Objective Functions (when objective_functions) are specified)

• Best Nonlinear Constraints

• Calibration (when calibration_terms are specified)

Theory

The basic steps of the soga algorithm are as follows:

1. Initialize the population

2. Evaluate the population (calculate the values of the objective function and constraints for each population member)

3. Loop until converged, or stopping criteria reached

1. Perform crossover

2. Perform mutation

3. Evaluate the new population

4. Assess the fitness of each member in the population

5. Replace the population with members selected to continue in the next generation

6. Test for convergence