A Study On Synchronization Of Fractional Order Hyperchaotic Based On Neural Network Sliding Mode Control

With the rise of information technologies such as big data,cloud computing,and Internet communication,people have attached great importance to the security of information.The chaotic secure communication can solve the problem of low efficiency of traditional encryption methods in encrypting large amounts of images and videos,and has higher real-time performance and security under the same resource occupancy conditions.In practical applications,the key to the realization of chaotic secure communication technology lies in the synchronization of chaotic systems.Fractional hyperchaotic systems have more positive Lyapunov exponents than general chaotic systems,and their chaotic attractor trajectories are separated in more directions,making them more secure in the field of chaotic secure communication.This also gives important significance to the research of synchronization of fractional hyperchaotic systems.Therefore,this paper mainly focuses on the synchronization of several types of fractional hyperchaotic systems,including the radial basis function(RBF)neural network sliding mode synchronization problem for fractional hyperchaotic systems with external disturbances,the dual neural network finite-time sliding mode synchronization problem for fractional hyperchaotic systems with internal uncertainties and external disturbances,and the type-2 fuzzy neural network fixed-time sliding mode synchronization problem for uncertain fractional hyperchaotic systems.The specific research content is as follows:Firstly,aiming at the synchronization problem of a class of fractional order hyperchaotic systems with unknown external disturbances,an adaptive sliding mode control scheme based on RBF neural networks is proposed.According to the Lyapunov stability theorem,the neural network weight adaptive law and switching gain adaptive law are derived.Through compensation control,the chattering problem of sliding mode controllers caused by unknown disturbances in the system is reduced,and complete synchronization between the drive system and the response system is ultimately achieved.Simulation results show that this control scheme has better rapidity and robustness than traditional single sliding mode controllers or single neural network controllers.Secondly,a novel dual neural network finite-time sliding mode control scheme is proposed for the synchronization of fractional order hyperchaotic systems with internal uncertainties and external disturbances.RBF neural network and recurrent neural network(RNN)adaptive laws are designed to approximate and compensate for internal uncertainties and external disturbances in the system.A new finite-time sliding mode surface is proposed.Based on finite-time stability theory and Lyapunov stability theorem,a finite-time sliding mode control algorithm is derived and designed.At the same time,differential evolution algorithm is used to optimize control parameters and switching gain,This greatly suppresses the chattering problem of the sliding mode controller caused by adaptive switching gain,and ultimately achieves fast synchronization between the drive system and the response system.Simulation results verify the effectiveness and superiority of this scheme.Finally,aiming at the synchronization problem of a class of uncertain fractional order hyperchaotic systems,a new type of self-evolving non-singleton interval type-2 probabilistic fuzzy neural network is proposed to approximate and compensate the deterministic and uncertain parts of the hyperchaotic system,and effectively suppress the fuzzy and random uncertainties in the control system.A new type of fixed-time sliding surface is designed,Based on the Lyapunov stability theorem and the proposed fuzzy neural network,a general fixed-time sliding mode control scheme with stable convergence time is proposed.Aiming at the "curse of dimensionality" problem of fuzzy neural networks,an improved whale optimization algorithm is proposed to optimize the neural network rule base,ultimately achieving complete synchronization between the drive system and the response system.Three sets of simulation results on different systems verify the effectiveness and superiority of this scheme.