net.sourceforge.cilib.problem
Class DeratingFunctionMaximisationProblem

java.lang.Object
  net.sourceforge.cilib.problem.OptimisationProblemAdapter
      net.sourceforge.cilib.problem.FunctionOptimisationProblem
          net.sourceforge.cilib.problem.FunctionMaximisationProblem
              net.sourceforge.cilib.problem.DeratingFunctionMaximisationProblem

All Implemented Interfaces:

Serializable, OptimisationProblem, Problem, Cloneable

public class DeratingFunctionMaximisationProblem
extends FunctionMaximisationProblem

Title: DeratingFunctionMaximisationProblem

Description: This class provides a way to modify the search space such that a solution could be removed from the search space and a new solution can be searched for.

The interested reader is referred to: Beasley, D., Bull, D. R. & Martin R. R. (1993). A Sequential Niche Technique for Multimodal Function Optimization. Evolutionary Computation 1(2), MIT Press, pp.101-125.

Author:

Clive Naicker

See Also:

Serialized Form

Field Summary

Fields inherited from class net.sourceforge.cilib.problem.FunctionOptimisationProblem

function

Fields inherited from class net.sourceforge.cilib.problem.OptimisationProblemAdapter

dataSetBuilder, fitnessEvaluations

Constructor Summary

DeratingFunctionMaximisationProblem()
The default constuctor.

Method Summary

void	addSolution(double[] solution) This method will add a solution to the vectorSolutions.
protected Fitness	calculateFitness(Type solution) Calculates the fitness of a solution with respect to the modifications to the search space.
void	clear() Removes all solutions from the vectorSolutions.
Function	getDeratingFunction() This method gets the function that is used to modify the search space, i.e.
double	getRawFitness(Type solution) This method finds the fitness of the solution without applying the derating function.
double	normalise(double distance) This method is used to calculate the normalised distance of a value in the domain of the problem.
void	setDeratingFunction(Function deratingFunction) This is an accessor method that can be used to set the deratingFunction that is used to modify the search space.
void	setFunction(Function function) This method sets the function to evaluate (that is the search space).

Methods inherited from class net.sourceforge.cilib.problem.FunctionMaximisationProblem

getClone, getError

Methods inherited from class net.sourceforge.cilib.problem.FunctionOptimisationProblem

accept, getBehaviouralDomain, getDomain, getFunction

Methods inherited from class net.sourceforge.cilib.problem.OptimisationProblemAdapter

changeEnvironment, getChangeStrategy, getDataSetBuilder, getFitness, getFitnessEvaluations, setChangeStrategy, setDataSetBuilder

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DeratingFunctionMaximisationProblem

public DeratingFunctionMaximisationProblem()

The default constuctor. Intialises the distanceMeasure to the EuclideanDistanceMeasure.

Method Detail

calculateFitness

protected Fitness calculateFitness(Type solution)

Calculates the fitness of a solution with respect to the modifications to the search space.

Overrides:

calculateFitness in class FunctionMaximisationProblem

Parameters:

solution - The solution to calculate the fitness.

Returns:

The fitness of the solution.

See Also:

OptimisationProblemAdapter.getFitness(Type, boolean)

getRawFitness

public double getRawFitness(Type solution)

This method finds the fitness of the solution without applying the derating function.

Parameters:

solution - The solution to find the fitness off.

Returns:

The fitness of the solution[]

addSolution

public void addSolution(double[] solution)

This method will add a solution to the vectorSolutions. The solution added will be used to create depression in the search using the deratingFunction.

Parameters:

solution - The position in the search to modify.

getDeratingFunction

public Function getDeratingFunction()

This method gets the function that is used to modify the search space, i.e. the derating function

Returns:

The deratingFunction

setDeratingFunction

public void setDeratingFunction(Function deratingFunction)

This is an accessor method that can be used to set the deratingFunction that is used to modify the search space.

Parameters:

deratingFunction - The derating function to use when modifying the search space.

normalise

public double normalise(double distance)

This method is used to calculate the normalised distance of a value in the domain of the problem.

Parameters:

distance - The distance to normalise

Returns:

A value in the domain [0..1]

clear

public void clear()

Removes all solutions from the vectorSolutions.

setFunction

public void setFunction(Function function)

This method sets the function to evaluate (that is the search space).

Overrides:

setFunction in class FunctionOptimisationProblem

Parameters:

function - The function to evaluate.