Design And Applications Of Fuzzy Logic Systems Based On Genetic Algorithm

In recent years,the development and application of fuzzy logic systems have become more and more extensive.It is undeniable that while fuzzy logic systems are being applied more and more,and the problems are getting better and better,the problems of fuzzy logic systems are becoming more and more obvious.Major fuzzy system identification problem.The fuzzy system identification problem includes structure identification and parameter identification.At present,the commonly used parameter adjustment schemes mainly include least squares method,GA algorithm,BP algorithm,particle group and so on.In this thesis,the type-1 TSK fuzzy logic system,interval type-2 TSK fuzzy logic systems and neural network are combined base on the IGA.The designed intelligent system is applied to the prediction of Mackey-Glass and international gold price,and the simulation research is given:(1)Introduce the relevant knowledge of the TSK fuzzy logic systems,neural network,GA and IGA,the mutation probability and the new IGA are proposed on the basis of GA,and the experimental results and advantages of IGA are given.(2)Research type-1 TSK fuzzy logic system based on IGA and give rules screening and parameter identification.The fuzzy logic system is integrated into the neural network to form a five-layer fuzzy neural network system.The IGA is used to filter the rules,then the IGA is used to optimize the system parameters.Finally the designed intelligent system model is applied to the prediction of Mackey-Glass and international gold prices and a comparison among the IGA,GA and BP algorithms are made.The simulation results show that the design of type-1 intelligent system is more feasible and effective.(3)On the basis of type-1 TSK fuzzy logic system,the design problem of interval type-2 TSK fuzzy logic system is studied.The interval type-2 TSK fuzzy logic system includes A1-C1,A2-C0 and A2-C1 interval type-2 TSK fuzzy logic systems.Each of A1-C1 and A2-C1 type Interval type-2 TSK fuzzy logic system design six fuzzy neural network systems,for A2-C0 interval type-2 TSK fuzzy logic system design five fuzzy neural network system.The IGA is used to filter the rules,then the IGA is used to optimize the system parameters.Finally the designed intelligent system model is applied to the prediction of Mackey-Glass and international gold prices and make a comparison among the IGA,GA and BP algorithms.The simulation results show that the interval type-2 TSK fuzzy logic system are more feasible and effective.(4)Comparing with the four types of TSK fuzzy logic systems,it can be known from the tracking effect graph and the root mean square error that the interval type-2 fuzzy logic system has a smaller error than the type-1 fuzzy logic system and has better precision.Compared with the three-interval TSK fuzzy logic system,the accuracy of A1-C1 fuzzy logic system and A2-C0 fuzzy logic system is not much different.However,the A2-C1 fuzzy logic system is the most accurate.The more parameters are visible,the more better the system performance.