net.sourceforge.cilib.pso.PSO Maven / Gradle / Ivy
/** __ __
* _____ _/ /_/ /_ Computational Intelligence Library (CIlib)
* / ___/ / / / __ \ (c) CIRG @ UP
* / /__/ / / / /_/ / http://cilib.net
* \___/_/_/_/_.___/
*/
package net.sourceforge.cilib.pso;
import com.google.common.collect.Lists;
import java.util.List;
import net.sourceforge.cilib.algorithm.initialisation.ClonedPopulationInitialisationStrategy;
import net.sourceforge.cilib.algorithm.population.IterationStrategy;
import net.sourceforge.cilib.algorithm.population.SinglePopulationBasedAlgorithm;
import net.sourceforge.cilib.entity.Topologies;
import net.sourceforge.cilib.entity.comparator.SocialBestFitnessComparator;
import net.sourceforge.cilib.problem.solution.OptimisationSolution;
import net.sourceforge.cilib.pso.iterationstrategies.SynchronousIterationStrategy;
import net.sourceforge.cilib.pso.particle.Particle;
import net.sourceforge.cilib.pso.particle.StandardParticle;
/**
*
* An implementation of the standard PSO algorithm.
*
*
* References:
*
*
*
* - J. Kennedy and R.C. Eberhart, "Particle swarm optimization," in Proceedings of IEEE
* International Conference on Neural Networks, vol. IV, (Perth Australia), pp. 1942-1948, 1995.
*
* - R.C. Eberhart and J. Kennedy, "A new optimizer using particle swarm theory," in Proceedings
* of the Sixth International Symposium on Micro Machine and Human Science, (Nagoya, Japan), pp.
* 39-43, 1995.
* - Y. Shi amd R.C. Eberhart, "A modified particle swarm optimizer," in Proceedings of the IEEE
* Congress on Evolutionary Computation, (Anchorage, Alaska), pp. 69-73, May 1998.
*
*
*/
public class PSO extends SinglePopulationBasedAlgorithm {
private static final long serialVersionUID = -8234345682394295357L;
private IterationStrategy iterationStrategy;
/**
* Creates a new instance of {@link PSO}.
*
* All fields are initialised to reasonable defaults. Note that the
* {@link net.sourceforge.cilib.problem.Problem} is initially {@code null}
* and must be set before {@link #algorithmInitialisation()} is called.
*/
public PSO() {
iterationStrategy = new SynchronousIterationStrategy();
initialisationStrategy = new ClonedPopulationInitialisationStrategy();
initialisationStrategy.setEntityType(new StandardParticle());
}
/**
* Create a copy of the provided instance.
* @param copy The instance to copy.
*/
public PSO(PSO copy) {
super(copy);
this.iterationStrategy = copy.iterationStrategy.getClone();
for (Particle p : topology) {
Particle nBest = Topologies.getNeighbourhoodBest(topology, p, this.neighbourhood, new SocialBestFitnessComparator());
p.setNeighbourhoodBest(nBest);
}
}
/**
* {@inheritDoc}
*/
@Override
public PSO getClone() {
return new PSO(this);
}
/**
* Perform the required initialisation for the algorithm. Create the particles and add then to
* the specified topology.
*/
@Override
public void algorithmInitialisation() {
this.topology = fj.data.List.iterableList(initialisationStrategy.initialise(optimisationProblem));
for (Particle p : topology) {
p.calculateFitness();
}
}
/**
* Perform the iteration of the PSO algorithm, use the appropriate IterationStrategy
* to perform the iteration.
*/
@Override
protected void algorithmIteration() {
iterationStrategy.performIteration(this);
}
/**
* Get the best current solution. This best solution is determined from the personal bests of the
* particles.
* @return The OptimisationSolution representing the best solution.
*/
@Override
public OptimisationSolution getBestSolution() {
Particle bestEntity = Topologies.getBestEntity(topology, new SocialBestFitnessComparator());
return new OptimisationSolution(bestEntity.getBestPosition(), bestEntity.getBestFitness());
}
/**
* Get the collection of best solutions. This result does not actually make sense in the normal
* PSO algorithm, but rather in a MultiObjective optimisation.
* @return The Collection<OptimisationSolution> containing the solutions.
*/
@Override
public List getSolutions() {
List solutions = Lists.newLinkedList();
for (Particle e : Topologies.getNeighbourhoodBestEntities(topology, neighbourhood, new SocialBestFitnessComparator())) {
solutions.add(new OptimisationSolution(e.getBestPosition(), e.getBestFitness()));
}
return solutions;
}
/**
* Get the IterationStrategy of the PSO algorithm.
* @return Returns the iterationStrategy..
*/
public IterationStrategy getIterationStrategy() {
return iterationStrategy;
}
/**
* Set the IterationStrategy to be used.
* @param iterationStrategy The iterationStrategy to set.
*/
public void setIterationStrategy(IterationStrategy iterationStrategy) {
this.iterationStrategy = iterationStrategy;
}
}