Research And Application Of A PSO Model Combining With GA

As an emerging field, since the 1980s swarm intelligence has attracted a number of researchers of many disciplines, and has become the hot spot of artificial intelligence as well as economics, sociology, biology and other interdisciplinary fields. Particle Swarm Algorithm that had been put forward in 1995 has also been successfully applied in many fields. Likely genetic algorithm, the initial population in PSO iterates step-by-step. There is not genetic operation(crossed operator and mutational operator) in PSO. Particles determine the course of search by their velocity. The mechanism of information sharing of PSO is different from that of GA, because particles in PSO have memory. Chromosomes share information mutually in GA, so the single swarm moves to optimal area evenly;whereas in PSO, merely the optimal particle shares its information with other particles. In PSO, it is a single-line flow of information, and the whole course of update follows the current optimal value,so all particles often converge to optimal value quickly. Owing to the simpleness of PSO, more and more researchers have done research on it, and it has become a hot spot of research.Because of the ever sharpness of the market competition, the expenditure idea has changed a lot. People no longer treat product function as the main factor of whether to buy something or not. Much attention has been paid to the attributes of creativity, outline, delightfulness as well as environment protection, which stand the important roll in competition. Facing this reality, many enterprises begin to realize the importance of product design and pay more attention to creative products. Creative design can satisfy the novel and ever-changing market demands and help to make the product more competitive effectively. Thus, creative design has been a popular research item attracts many researchers in relevant fields home and abroad.By far, no systematic method can direct designers to carry out creative design orderly. In fact, most of the instances are created by modifying the products already existed rather than design without any foundation. The rise of the evolution computation provides us a new approach for conceptual creative design; therefore, it has become one of the most important techniques for computational and creative design.Much attention has been paid to the construction and application of an improved particle swarm optimizational model combining with genetic algorithm in this paper. So that it can apply evolution design to new intelligent algorithms and provide a supporting platform of creation for distributed designers by studying a new optimizational mechanism especially for creative design.The main work is as follows:1. Propose an evolutional mechanism based on Cloud Model Cloud model expressed by lingual value is an uncertain transforming model between qualitative concept and their quantitative expressions, and it combines fuzziness with randomicity of uncertain concept organicly. Owing to the uncertainty of conceptual designing, an improved genetic algorithm based on cloud model is presented in this paper. The results show that it can preserve the ability of global optimization and overcome the disadvantage of the original genetic algorithm, quicken the speed of design and widen the thought of design, which improve the innovation of component conceptual design.2. Put forward an improved particle swarm optimizational model based on dynamic nicheThe niche provide probability of shaping of new species, and it is one of the ultimate reasons of infinite diversity of nature. Existing particle swarm algorithm and its variations converge slowly and get into local extremum easily. Due to this, based on the characteristics of current conceptual designing, an improved particle swarm optimizational model based on dynamic niche is proposed in this paper.3. Bring forward a particle swarm optimizational technique combining with GABased on the dynamic niche PSO algorithm, this paper introduces the mechanism of crossover and mutation in order to make swarms have genetic characteristics. Hybridization can make full use of existing information, whereas mutation can generate new solutions to expand the scope of search. Hybridization after the emergence of new solution can produce a better solution; when the search get into stagnancy , swarms implement mutation. Hybridization and mutation can be complementary, and combining them can achieve the complementary advantages, thereby reduce the computing time and avoid premature.4. Implement a mobilephone creative design system based on the PSO combining with GAThe improved PSO model combining with GA is applied to mobilephone design field and a design system, on the base of the model, is realized. With entity modeling engine ACIS utilized as the sculpt kernel, the system is designed on WindowsXP platform with VC++.NET 2003. In addition, SQL Server 2000 is used as database system. Finally, the analysis of experiment shows that the design results are quite satisfying which is based on the design, selection and assembly of mobilephone.