Title :
Parameter approximate dynamic optimization for PSO systems
Author :
Kang, Qi ; Wang, Lei ; Liu, Derong ; Wu, Qidi
Author_Institution :
Key Lab. of Embedded Syst. & Comput.-Service of MOE, Tongji Univ., Shanghai, China
Abstract :
This paper presents a novel swarm approximate dynamic programming method (swarm-ADP) for parameter optimization of PSO systems, from the perspective of optimal control. Based on the stability results of a simplified PSO and the swarm-ADP algorithm, parameter dynamic optimization and computation is studied in detail for a deterministic PSO system and a stochastic PSO system, respectively. Furthermore, numerical simulations based on several benchmarks optimization are performed. The results show the validity of the proposed parameter optimization method for PSO systems.
Keywords :
dynamic programming; optimal control; particle swarm optimisation; stability; stochastic systems; PSO systems; deterministic PSO system; optimal control; parameter approximate dynamic optimization; stability results; stochastic PSO system; swarm approximate dynamic programming method; Adaptive control; Approximation algorithms; Estimation error; Multidimensional systems; Signal processing algorithms; Software algorithms; Stochastic processes; Stochastic resonance; Stochastic systems; Upper bound;
Conference_Titel :
Decision and Control, 2009 held jointly with the 2009 28th Chinese Control Conference. CDC/CCC 2009. Proceedings of the 48th IEEE Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-3871-6
Electronic_ISBN :
0191-2216
DOI :
10.1109/CDC.2009.5400841
