Interval Type-2TSK Fuzzy Logic System Design And Application Research

In recent years, fuzzy logic system has become one of the hot research topics in the fieldof academic and practical application. Especially TSK fuzzy logic system can show thesuperior model and control ability in complex, nonlinear and multivariable systems. Soresearch the TSK fuzzy logic system has the practical and realistic significance. The thesismainly researches the design of the type-1TSK fuzzy logic system and the interval type-2TSK fuzzy logic system. And the design of TSK fuzzy logic system is applied to corn borerforecasts and the Shanghai Composite Index closing price forecasts, given the simulationresearch. Main works are summarized as follows:(1) Introduce the development and application of fuzzy logic system and introduce somerelevant knowledge of algorithms.(2) Introduce the structure of the type-1fuzzy logic system and the interval type-2fuzzylogic system.(3) Research the design of the type-1TSK fuzzy logic system. Using fuzzy C-meansclustering method filter rules, given the antecedent parameters, confirm the consequentparameters by least square method. Then the system is applied to forecast the corn borer andsimulation with MATLAB. The simulation results show that the designed type-1TSK fuzzylogic system is effective and feasible.(4) Research the design of the interval type-2TSK fuzzy logic system. Design threekinds of interval type-2TSK fuzzy logic systems. They are A2-C1, A2-C0and A1-C1. Usingthe clustering analysis filtering rules, using BP algorithm adjust the antecedent andconsequent parameters of the systems.(5) Forecast Shanghai (securities) composite index closing price using the designed threekinds of interval type-2TSK fuzzy logic systems, and simulation with MATLAB. Thesimulation results show that the designed three kinds of interval type-2TSK fuzzy logicsystems are effective and feasible. The tracking effect and root mean square errors can showthat interval type-2TSK fuzzy logic system than type-1TSK fuzzy logic system has betterstability and accuracy. Compared the three kinds of interval type-2TSK fuzzy logic systemwe can know the more design parameters the more accurate of the system.