Study on the Strategy of Decreasing Inertia Weight in Particle Swarm Optimi-ation Algorithm [J]. 2006, 40(1): 53-0056. DOI: 10.7652/xjtuxb200601013.
DOI:
Study on the Strategy of Decreasing Inertia Weight in Particle Swarm Optimi-ation Algorithm [J]. 2006, 40(1): 53-0056. DOI: 10.7652/xjtuxb200601013.DOI:
Study on the Strategy of Decreasing Inertia Weight in Particle Swarm Optimi-ation Algorithm
To efficiently cont rol t he global and local search of particle swarm optimi-ation ( PSO)
<br />motivated by t he idea of decreasing inertia weight (DIW)
t hree nonlinear st rategies for DIW
a<br />parabola opening upwards
a parabola opening downwards and an exponential curve
are pro2<br />posed based on the existing linear DIW. Sp here
Rosenbrock
Griewank and Rast rigrin f unctions<br />are used to evaluate t he st rategies on t he PSO performance. The experimental result s show t hat<br />for most continuous optimi-ation problems
t he st rategy of concave f unction gains an advantage<br />over t he linear st rategy
while the linear st rategy outperforms st rategy of convex f unction wit h<br />t he identical initial and final weight s.<br />
关键词
Keywords
references
Chen S J , Lin L. Decomposition of interdependent task
group for concurrent engineering [ J ] . Computers and