net.sourceforge.cilib.pso.velocityupdatestrategies
Class ConstrictionVelocityUpdate

java.lang.Object
  net.sourceforge.cilib.pso.velocityupdatestrategies.ConstrictionVelocityUpdate

All Implemented Interfaces:

Serializable, VelocityUpdateStrategy, Cloneable

public class ConstrictionVelocityUpdate
extends Object
implements VelocityUpdateStrategy

A velocity update strategy that utilizes the constriction coefficient as developed by Clerc.

References:

• @INPROCEEDINGS{785513,
  title={The swarm and the queen: towards a deterministic and adaptive particle swarm optimization},
  author={Clerc, M.},
  booktitle={Evolutionary Computation, 1999. CEC 99. Proceedings of the 1999 Congress on},
  year={1999},
  month={},
  volume={3},
  number={},
  pages={-1957 Vol. 3},
  abstract={A very simple particle swarm optimization iterative algorithm is presented, with just
  one equation and one social/confidence parameter. We define a “no-hope” convergence criterion and
  a “rehope” method so that, from time to time, the swarm re-initializes its position, according to
  some gradient estimations of the objective function and to the previous re-initialization (it means
  it has a kind of very rudimentary memory). We then study two different cases, a quite “easy” one
  (the Alpine function) and a “difficult” one (the Banana function), but both just in dimension two.
  The process is improved by taking into account the swarm gravity center (the “queen”) and the results
  are good enough so that it is certainly worthwhile trying the method on more complex problems},
  keywords={adaptive systems, deterministic algorithms, evolutionary computation, iterative methodsAlpine function,
  Banana function, adaptive particle swarm optimization, gradient estimations, no-hope convergence criterion,
  objective function, queen, re-initialization, rehope method, rudimentary memory, simple particle swarm
  optimization iterative algorithm, social/confidence parameter, swarm gravity center},
  doi={10.1109/CEC.1999.785513},
  ISSN={}, }
  

• @ARTICLE{985692,
  title={The particle swarm - explosion, stability, and convergence in a multidimensional complex space},
  author={Clerc, M. and Kennedy, J.},
  journal={Evolutionary Computation, IEEE Transactions on},
  year={2002},
  month={Feb},
  volume={6},
  number={1},
  pages={58-73},
  abstract={The particle swarm is an algorithm for finding optimal regions of complex search spaces
  through the interaction of individuals in a population of particles. This paper analyzes a particle's
  trajectory as it moves in discrete time (the algebraic view), then progresses to the view of it in
  continuous time (the analytical view). A five-dimensional depiction is developed, which describes
  the system completely. These analyses lead to a generalized model of the algorithm, containing a set
  of coefficients to control the system's convergence tendencies. Some results of the particle swarm optimizer,
  implementing modifications derived from the analysis, suggest methods for altering the original algorithm
  in ways that eliminate problems and increase the ability of the particle swarm to find optima of some well-studied
  test functions },
  keywords={convergence of numerical methods, genetic algorithms, numerical stability, search problemsconvergence,
  evolutionary computation, multidimensional complex space, optimization, particle swarm, particle trajectory,
  search spaces, stability},
  doi={10.1109/4235.985692},
  ISSN={1089-778X}, }
  

Note, this strategy does not the inertia control parameter. Certain constraints are imposed on the other control parameters in order to calculate the constriction coefficient, namely: $c1r1 + c2r2 \leq 4$ , and $\kappa \in [0, 1]$

Author:

andrich

See Also:

Serialized Form

Constructor Summary

ConstrictionVelocityUpdate()
Default constructor.

ConstrictionVelocityUpdate(ConstrictionVelocityUpdate copy)
Copy constructor.

Method Summary

protected void	clamp(Vector velocity, int i) Clamp to maximum velocity.
ConstrictionVelocityUpdate	getClone() Clone the VelocityUpdateStrategy object.
ControlParameter	getCognitiveAcceleration() Get the coginitive acceleration parameter.
ControlParameter	getKappa() Get the Kappa control parameter.
ControlParameter	getSocialAcceleration() Get the social acceleration parameter.
ControlParameter	getVMax() Get the maximum velocity parameter.
void	setCognitiveAcceleration(ControlParameter cognitiveAcceleration) Set the coginitive acceleration parameter.
void	setKappa(ControlParameter kappa) Set the Kappa control parameter.
void	setSocialAcceleration(ControlParameter socialAcceleration) Set the social acceleration parameter.
void	setVMax(ControlParameter vMax) Set the maximum velocity parameter.
void	updateControlParameters(Particle particle) Update the needed control parameters for the VelocityUpdate, if needed.
void	updateVelocity(Particle particle) Perform the velocity update operation on the specified Particle.

Methods inherited from class java.lang.Object

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

Constructor Detail

ConstrictionVelocityUpdate

public ConstrictionVelocityUpdate()

Default constructor. The values given to the control parameters attempt to adhere to the constraints of calculating the constriction constant, but do not necessarily represent good values.

ConstrictionVelocityUpdate

public ConstrictionVelocityUpdate(ConstrictionVelocityUpdate copy)

Copy constructor.

Parameters:

copy - the ConstrictionVelocityUpdate to copy.

Method Detail

getClone

public ConstrictionVelocityUpdate getClone()

Clone the VelocityUpdateStrategy object.

Specified by:

getClone in interface VelocityUpdateStrategy

Specified by:

getClone in interface Cloneable

Returns:

A cloned VelocityUpdateStrategy

See Also:

Object.clone()

updateVelocity

public void updateVelocity(Particle particle)

Perform the velocity update operation on the specified Particle.

Specified by:

updateVelocity in interface VelocityUpdateStrategy

Parameters:

particle - The Particle to apply the operation on.

clamp

protected void clamp(Vector velocity,
                     int i)

Clamp to maximum velocity.

Parameters:

velocity - The Vector to be clamped.

i - The dimension index to be clamped

updateControlParameters

public void updateControlParameters(Particle particle)

Update the needed control parameters for the VelocityUpdate, if needed.

Specified by:

updateControlParameters in interface VelocityUpdateStrategy

Parameters:

particle - The particle for whom the VelocityUpdateStrategy parameters need to be updated.

getCognitiveAcceleration

public ControlParameter getCognitiveAcceleration()

Get the coginitive acceleration parameter.

Returns:

the cognitive acceleration control parameter .

setCognitiveAcceleration

public void setCognitiveAcceleration(ControlParameter cognitiveAcceleration)

Set the coginitive acceleration parameter.

Parameters:

cognitiveAcceleration - the new cognitive acceleration control parameter .

getKappa

public ControlParameter getKappa()

Get the Kappa control parameter.

Returns:

the kappa control parameter .

setKappa

public void setKappa(ControlParameter kappa)

Set the Kappa control parameter.

Parameters:

kappa - the new kappa control parameter .

getSocialAcceleration

public ControlParameter getSocialAcceleration()

Get the social acceleration parameter.

Returns:

the social acceleration control parameter .

setSocialAcceleration

public void setSocialAcceleration(ControlParameter socialAcceleration)

Set the social acceleration parameter.

Parameters:

socialAcceleration - the new social accerelation control parameter .

getVMax

public ControlParameter getVMax()

Get the maximum velocity parameter.

Returns:

the maximum velocity control parameter .

setVMax

public void setVMax(ControlParameter vMax)

Set the maximum velocity parameter.

Parameters:

vMax - the new maximum velocity control parameter .