mopsocd (version 0.5.1)

mopsocd: MOPSOCD: Multi-objective Particle Swarm Optimization with Crowding Distance

Description

Multi-objective optimization involves maximizing or minimizing multiple interacting and/or conflicting objective functions subject to a set of contraints. MOPSOCD is a multi-objective optimization solver based on particle swarm optimization that uses crowding distance computation to ensure an even spread of non-dominated solutions.

Usage

mopsocd(fn, gn, varcnt, fncnt, lowerbound, upperbound, opt, popsize, maxgen, archivesize, verbosity, pMut, w, c1, c2)

Arguments

fn

Objective functions to be optimized

gn

Constraints (optional)

varcnt

Number of Parameters

fncnt

Number of Objectives

lowerbound

Parameter Lower Bound

upperbound

Parameter Upper Bound

opt

Optimization type (0: minimization; 1: maximization)

popsize

Population Size (default: 100)

maxgen

Number of Generations (default: 100)

archivesize

Maximum size of archive containing non-dominated points (default: 250)

verbosity

Verbosity Levels : 0,1,2,3 (default: 1)

pMut

Mutation Probability (default: 0.5)

w

Inertia Weight (default: 0.4)

c1

Acceleration Coefficient 1 (default: 1.0)

c2

Acceleration Coefficient 2 (default: 1.0)

Value

The returned value is a pareto object with the following fields:

numsols

Number of Solutions Found

paramvalues

Estimated Parameter Values

objfnvalues

Values of the Objectives

References

C. R. Raquel and P.C. Naval, "An Effective use of Crowding Distance in Multiobjective Particle Swarm Optimization", Proceedings of Genetic and Evolutionary Computation Conference (GECCO 2005), Washington, D.C., June 25-29, 2005.

See Also

examples pareto