The Construction Of Mamdani Fuzzy System And The Hybrid Algorithm Of Polyline Fuzzy Neural Network

The generation of t.he fuzzy system is made up of the fuzzy rule base?inference engine?fuzzifier and defuzzifier.Its main function is to deal with incomplete data and language information at the same time,its essence is a mapping from the input domain to the output domain,and the fuzzy rule base is composed of several fuzzy IF-THEN rules.Polyline fuzzy neural network completes the fuzzy information processing by limited points of fuzzy number,its expression is an arithmetic operation system based on the polyline fuzzy number.Although.the two models are not depend on the exact mathematical model,it has a strong ability in the approximation of nonlinear functions.Chaotic mapping is a universal nonlinear process in nature,it has been widely used in the fields of image encryption,communication technology and function optimization,and its main feature is that it is very sensitive to the initial value.Therefore.it is of great theoretical and practical significance to design the optimization algorithm of the broken line fuzzy neural network by using the principle of chaos mapping.In the first part:Introduction,introduces the background and research status of this topics.In the second part:Preparatory knowledge,we introduce the line fuzzy number and its operationwe,Gaussian(Triangular)fuzzifier,Mamdani fuzzy system and chaotic random sequence and other related knowledge.separately.In the third part:When the fuzzy rule precursor is tricangular fuzzy number.we give the mem-bership function of the antecedent fuzzy sets by the function extremum method then the analytic representation of two Mamdani fuzzy systems based on Triangular fuzzifier and Gauss fuzzifier is given.In addition,a chaotic genetic algorithm for single input and single output Mamdani fuzzy systems is designed based on random sequences generated by chaotic maps.and the validity of the algorithm is illustrated by the prediction curve of the Machey-Glass chaotic time series.In the final part:Firstly,random sequences are constructed by a chaos random generator.These random sequences are used as the initial population of chaos GA.Then the optimal indi-vidual based on the chaos search method as initial population of particle swarm optimization,and then propose the CGA-PSO hybrid optimization algorithm for the polyline Fuzzy Neural Network.At last.through simulation examples that the CGA-PSO hybrid algorithm avoids the dependence of the traditional genetic algorithm on the initial value.and the algorithm has the advantages of fast convergence and good stability.