Design And Applications Of Fuzzy Logic Systems Based On QPSO Intelligence Algorithm

Due to its wide application,fuzzy logic system has become a hot research topic in the field of academic and practical application.Fuzzy system identification includes two aspects of structural identification and parameter identification.For parameter identification,the current method often uses the least square method,BP algorithm,etc.Although there are a lot of methods,but it is not intelligent algorithm.In this thesis,the type-1 TSK fuzzy logic system?interval type-2 TSK fuzzy logic systems A1-C1?A2-C0?A2-C1(where A-antecedent parameters,and C-consequent parameters)and neural network are combined,four kinds of fuzzy neural network system were designed,and the quantum behaved particle swarm optimization(QPSO)intelligent algorithm use to adjust the fuzzy logic system parameters,the design of intelligent system is applied in the prediction of the International Gold Prices and NASDAQ Composite Index,and the simulation is given.The specific work is as follows:(1)Introduce the relevant knowledge of the TSK fuzzy logic systems?neural network and QPSO algorithm.(2)Research type-1 TSK fuzzy logic system based on QPSO algorithm.The fuzzy logic system is integrated into the neural network to get the five layer fuzzy neural network system.QPSO algorithm is used to filtering rules.And then apply the QPSO algorithm to optimize the system parameters.Finally,the design of intelligent system model is applied in the prediction of the International Gold Prices and NASDAQ Composite Index.Then make a comparison between the QPSO algorithm and BP algorithm.The simulation results show that the design of type-1 intelligent system is feasible and effective.(3)On the basis of type-1 TSK fuzzy logic system,the antecedent is invariant,the consequent is changed into type-1 fuzzy set,and the A1-C1 interval type-2 TSK fuzzy logic system is obtained.The fuzzy logic system is integrated into the neural network to get the six layer fuzzy neural network system.QPSO algorithm is used to filtering rules.And then apply the QPSO algorithm to optimize the system parameters.Finally,the design of intelligent system model is applied in the prediction of NASDAQ Composite Index.Then make a comparison between the QPSO algorithm and BP algorithm.The simulation results show that the design of A1-C1 intelligent system is feasible and effective.(4)On the basis of type-1 TSK fuzzy logic system,the consequent is invariant,the antecedent is changed into interval type-2 fuzzy set,and the A2-C0 interval type-2 TSK fuzzy logic system is obtained.The fuzzy logic system is integrated into the neural network to get the five layer fuzzy neural network system.QPSO algorithm is used to filtering rules.And then apply the QPSO algorithm to optimize the system parameters.Finally,the design of intelligent system model is applied in the prediction of NASDAQ Composite Index.Then make a comparison between the QPSO algorithm and BP algorithm.The simulation results show that the design of A2-C0 intelligent system is feasible and effective.(5)Combined(3)and(4),the antecedent is type-2 fuzzy set,the consequent is type-1fuzzy set,and the A2-C1 interval type-2 TSK fuzzy logic system is obtained.The fuzzy logic system is integrated into the neural network to get the six layer fuzzy neural network system.QPSO algorithm is used to filtering rules.And then apply the QPSO algorithm to optimize the system parameters.Finally,the design of intelligent system model is applied in the prediction of NASDAQ Composite Index.Then make a comparison between the QPSO algorithm and BP algorithm.The simulation results show that the design of A2-C1 intelligent system is feasible and effective.Compare the four kinds of TSK fuzzy logic system,though the tracking diagrams and the root mean square error that can be seen the error of the interval type-2 TSK fuzzy logic system is relatively smaller than type-1 TSK fuzzy logic system,and has a better precision.