Composite Learning Sliding Mode Synchronization Of Chaotic Fractional-order Neural Networks

Sliding mode control is one of the most effective mathematical methods to study the synchronization of chaotic fractional-order neural networks.Chaotic fractionalorder neural networks are easily affected by uncertainties and external disturbances,and the synchronization of chaotic fractional-order systems with mismatched parameters is more complex.In this paper,sliding mode control and composite learning sliding mode control are proposed to study the synchronization of chaotic fractionalorder neural networks.However,traditional sliding mode control can only realize the synchronization of chaotic fractional-order neural networks under the condition of permanent stimulus.In order to overcome this limitation,online recorded data and instantaneous data are used to define the prediction error of uncertain parameters,and the synchronization error and the prediction error are combined to construct the composite learning law.The proposed composite learning sliding mode control can not only guarantee the synchronization of chaotic fractional-order neural networks under the condition of interval stimulation,but also realize the accurate estimation of mismatched parameters.This paper contains three chapters,the details are as follows:Chapter 1:Preliminaries.First,this chapter introduces the definition of fractional calculus and some lemmas.Secondly,it describes the problems to be solved in this paper.Chapter 2:Adaptive sliding mode control and composite learning sliding mode control of chaotic fractional-order neural networks.First,the sliding mode control is constructed and the stability of chaotic fractional-order neural network is analyzed.Secondly,the composite learning sliding mode control is designed and the stability of chaotic fractional-order neural network is analyzed to make chaotic fractional-order neural network synchronizes and accurately estimates the parameters.Chapter 3:Comparison and analysis of the control performance of sliding mode control and composite learning sliding mode control.A simulation example is given to test the control effect of the designed sliding mode control method and composite learning sliding mode control method.Secondly,the control performance of the two methods axe compared and analyzed.