Location:Home > Engineering science > Computer Science > Computer technology > Research and Application on the Hybrid Algorithm of PSO And ABC
Details
Name

Research and Application on the Hybrid Algorithm of PSO And ABC

Downloads: []
Author
Tutor: SunHui
School: Nanchang University of Aeronautics and
Course: Computer technology
Keywords: Particle Swarm Optimization Algorithm,Artificial Bee ColonyAlgorithm,Hybrid algo
CLC: TP18
Type: Master's thesis
Year:  2013
Facebook Google+ Email Gmail Evernote LinkedIn Twitter Addthis

not access Image Error Other errors

Abstract:
Optimization problem is a hot topic in research, which existing in every field of human society. Swarm intelligence optimization algorithm is a new type of optimizationalgorithm. Particle Swarm Optimization(PSO)simulates the foraging behavior ofbirds to find the optimal solution. It is a widely used algorithm in several intelligentalgorithms. It can quickly converge to the optimal solution for unimodal functions.Artificial Bee Colony algorithm is an optimization algorithm based on simulating theforaging behavior of honey bee swarm, can get better performance in multi-peakfunctions by exploring better nectar source than the current positions. In this paper,three improved optimization algorithms are proposed according to the characteristicsof PSO algorithm and ABC algorithm, and one of improved algorithm is applied tothe optimization problem of wireless sensor network coverage.(1) This paper proposes a new PSO algorithm with fast convergence rate andadaptive escape function(FAPSO). The FAPSO algorithm adds local search strategywhich is from ABC algorithm, and particles search two times including a globalsearch and a local search in every iteration. The FAPSO also adds the function ofescape, which is learns from that the scout bee in theABC algorithm can jump out oflocal optimum. Simulation results show that the FAPSO algorithm has good globalsearch performance,and increase the convergent speed. Moreover, this algorithm caneffectively avoid the premature phenomenon.(2) TheABC and PSO adaptive interactive learning optimization algorithm(ABC-PSO) is proposed in this paper. TheABC-PSO algorithm divides one groupinto two subgroups. One subgroup is evolved by the improvedABC algorithm, andanother subgroup is evolved by the PSO algorithm.At the same time, the twosubgroups select different learning strategies adaptively and learn the usefulinformation each other. Simulation results show that theABC-PSO algorithm whichcombines the advantages ofABC and PSO balances the local exploitation ability andglobal exploration ability while keep the population diversity.(3) A mixed optimization algorithm with the individuals of ABC and PSO isproposed and applied in wireless sensor network. In this algorithm, the swarm isdivided into two subgroups. One sub-group evolves using ABC algorithm, the othersub-group evolves using PSO. The information exchange of two subgroups can become true through sharing the mutual one. The optimal value of PSO is replaced by optimum value of ABC at the appropriate time. It can guide the individual to move tothe better position. Through worst position of employed bee replacing by the optimalvalue of PSO, it can attract more onlookers move to good source position. Theexperimental results show that the PABC algorithm has better optimal precisionand performance than ABCand PSO algorithm on the high and low dimension.Bettercover the optimization effect is achieved through the PABC algorithm in wirelesssensor networks.
Related Dissertations
Last updated
Sponsored Links
Home |About Us| Contact Us| Feedback| Privacy | copyright | Back to top