Research And Application Of Intelligence Algorithms In Fuzzy Logic And Fuzzy System

Intellegence provides a new way with which it is possible to explore collective problem solving without centralized control or the provision of a global model.It can solve the specific problems by simulating the collective behaviors. The characteristic of Intelligence algorithm is stochastic, parallel and distributed.Because of its character, intelligence algorithm is widely used in many aspects.Recently, the theory of fuzzy logic and fuzzy system has made great progress. The success realization of fuzzy logic and fuzzy system in life and industry control makes more and more scholars do deeper research into the theory. The combination with the intelligence algorithm and the theory about fuzzy logic and fuzzy system let model be more effective and efficiency, also fuzzy logic and fuzzy system can take advantage of the theory about fuzzy to make it more fit to apply to manufacture.Theoretical analyses and algorithm improving on intelligence algorithm are mainly discussed in our work and the application of fuzzy logic and fuzzy system is also studied in the work.The main contents of this dissertation are as follows:(1)Research background of intelligence algorithm and the theory about fuzzy logic and fuzzy system is explained.The current situation of algorithms is detailed introduced.The research method in fuzzy logic and fuzzy system is presented. Research methods nad ideas in the work are proposed.(2)The work systematically introduce idea and way of intelligence algorithm, focuses on Particle Swarm Optimization Algorithm (PSO).Convergence of PSO algorithm is analyzed by algebraic method and then the conditions of convergence and repulse for PSO algorithm are reached. Premature convergence is also appeared in PSO algorithm when solving multimodal problems. According to the conclusion, improvement is made by providing many ways.(3)Because of PSO's shortcomings that can't convergence at possibility of 100 percent, premature convergence and so on, Particle Swarm Optimization Algorithm based on Quantum-behaved behavior is presented.Aimed at local convergence in QPSO algorithm when solving multimodal problems, the reason for local convergence lies in the collections of swarm diversity decline and the particles lose the ability of searching in a wide space.The improvement is realized by mutating the dimension and taking on the technique of nache and charotic mutation. The improved QPSO algorithm shows preferable ability in solving the multimodal problems.(4)Fuzzy Cognitive Maps (FCMs) is a kind of soft computing tools, which is combinatory about fuzzy (FNN) and fuzzy logic. Owe to the intuitive expression ability and powerful inference ability, FCMs is more semantic than FNN. So FCMs is more suitable to research into the complex system because of the characterstic——more interpretability for model built.Meanwhile, the theory about maps is introduced and can be done deeply research into FCMs in the characteristic of maps with directions. In view of the advantages of FCMs, it can be successfully introduced into industry control model and specific language imparement clinic system, which both make great progress. (5) 2D Otsu method, which considers the gray information and spatial neighbor information between pixels inthe image simultaneously, is an efficient image segmentation method. However, the computational burden of finding optimal threshold vector is very large for 2D Otsu method. An optimization method, i.e., quantum-behaved particle swarm optimization (QPSO), is used to find the best 2D threshold vector, Experimental result s show that the proposed method can not only obtain ideal segmentation results but also decrease the computation cost reasonably.The work proposed a method of Median Filter Based on Fuzzy Theory by QPSO Algorithm Optimized to filter the noise in image.Using our new algorithm to filter the pulse noise, a better result can be got.(6)Almost all systems in real life are non-lined, so it is necessary to do research into the recognition of the nonlinear system. Takagi-Sugeno fuzzy model can approach the Precision sets with any precision.A new identification algorithm ofT akagi-S ugeno fuzzy modeli s proposed.An adaptive fuzzy clustering algorithm is applied to decide prefix construction and parameters of fussy model.Quantuam Particle Swarm Optimization (QPSO) algorithm is used to get the result parameters of fussy model, which can gain optimal values of system parameters.The simulation results show that the method is effective.The main contributions of this paper are summarized and the further researches on work are suggested at the end of this dissertation.