Font Size: a A A

The Research Of Particle Swarm Optimization Based On Multi-factor Inertia Weight

Posted on:2014-12-16Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZhangFull Text:PDF
GTID:2308330452956815Subject:Software engineering
Abstract/Summary:PDF Full Text Request
The concept of artificial intelligence was proposed in the middle1950s. Then, theresearch and exploration of artificial intelligence attracts a great many excellentinformation experts, mathematicians and other scholars. While the research of artificialintelligence becomes more and more mature, intelligent algorithm is born. Thedevelopment of intelligent algorithm is very rapid, and various optimization problems aresolved efficiently by intelligent algorithms. Swarm intelligent algorithm, which isproposed in the1990s, has become the key point of intelligent algorithm.Particle swarm optimization (PSO) is a stochastic population-based algorithmmotivated by intelligent collective behavior of some animals. Comparing with otherintelligence optimizations, the most important advantages of the PSO are that PSO is easyto implement and there are few parameters to adjust. For the reasons above, PSOalgorithm has attracted more and more researchers. And the researchers devote themselvesto improve the PSO algorithm. In order to improve the PSO algorithm, we discuss themeaning of optimization problem and the concept of intelligent optimization algorithm.Then, we present some related intelligent optimization algorithms, principles and theirbasic flows. After that, we talk about the occurrence of PSO algorithm, and introduce thewhole framework of PSO’s mathematical model. And we analyze every component of theframework of PSO algorithm and their effect on the algorithm’s performance carefully.Then, we summarize various methods that can improve the performance of PSO algorithm.After the above, we take the strategy of inertia weight as breakthrough point, reviewvarieties of existing inertia weight strategies, and absorb the experience. Then, deliberatethe use of feedback information of different inertia weight strategies, and put forward theidea of pluralism and layering based on the deliberation. After that, PSO algorithm basedon multi-factor inertia weight is proposed by the instruction of the idea above. Then,a series of benchmark functions are used in the experiment, which demonstrate that thenew strategy is more effective than the previous ones. At last, the PSO based onmulti-factor adaptive inertia weight is used to optimize a real world problem of pressurevessel design.
Keywords/Search Tags:Intelligent optimization, Particle swarm optimization algorithm, Inertia weight, Pressure vessel design
PDF Full Text Request
Related items