Font Size: a A A

The Research Of Adaptive Parameter Selection And Neighborhood Topology Based On Quantum-Behaved Particle Swarm Optimization

Posted on:2009-12-17Degree:MasterType:Thesis
Country:ChinaCandidate:L D KongFull Text:PDF
GTID:2178360272956852Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
The purpose of the paper is to research and improve Quantum-behaved Particle Swarm Optimization(QPSO) algorithm. Swarm intelligence algorithm belongs to evolutionary genus algorithm, which is a an effective method of solving optimization problem, especially for complex system. QPSO is a recently proposed swarm intelligence algorithm with global convergence and its performance precedes Particle Swarm Optimization(PSO) and Genetic Algorithm far, which is shown by some previous work. Thus, the research of this paper is of great scientific significance and research value in evolutionary computation area.Particle Swarm Optimization(PSO), a swarm intelligence algorithm in this field is described first. A new global convergence algorithm, Quantum-behaved Particle Swarm Optimization (QPSO) algorithm is mainly formulated. Then, the two algorithms, PSO and QPSO are compared with each other, analysis and discussion of their merit and demerit is given.Two improved Quantum-behaved Particle Swarm Optimization algorithms based on QPSO are presented, which are called Adaptive Quantum-behaved Particle Swarm Optimization on Global Level (AQPSO) and Quantum-behaved Particle Swarm Optimization with Neighbourhood Operator(NQPSO). AQPSO suggests a parameter control method based on the global level and is introduced a diversity-guided model into the QPSO to make the PSO system an open evolutionary particle swarm which improves the global searh ability of the algorithm. NQPSO is introduced a concept of neighbourhood topology into the QPSO to ensure the diversity of the swarm, in which each particle only can communicate information with the particle in its neighborhood. The performance of the two improved algorithms is compared with those of Standard PSO (SPSO) and original QPSO by testing the algorithms on several benchmark functions. The experiments results show that the improved algorithms outperforms due to its strong global search ability, particularly in the optimization problems with high dimension.Finally, the modified QPSO algorithm is used to solve traveling salesman problem(TSP), and the result shows that it can find optimum solutions more effectively in quantily than SPSO or QPSO.
Keywords/Search Tags:particle swarm optimization, quantum-behaved, diversity-guided model, daptive, neighbourhood topology, traveling salesman problem
PDF Full Text Request
Related items