Stability Analysis For Neural Networks Of Neutral Type With Semi-markovian Jump Parameters

This paper probes into the synchronization for memristor-based hybrid neural networks via nonlinear coupling and the stochastic stability for stochastic neural networks of neutral type with semi-Markov jump parameters.And different Lyapunov-Krasovskii functionals are constructing respectively,using the technique of linear matrix inequality to make the considered neural network reach a stable state.Chapter 1 mainly introduces the research status and background of memristive neural network,hybrid neural network,neural networks of neutral type and semi-Markov jumping system.Also,the framework of the research structure of this paper is given.Chapter 2 considers the nonlinear coupling synchronization problem of memristive neural network.First,the purpose is to confirm whether the quadratic function is negative on the closed interval,not care the concavity of the function.Then,Based on Legendre polynomials,a recent extended integral inequality with free matrices is popularized to get tighter lower bound of some integral terms.Next,planning a Lyapunov functional,by applying the new integral inequality with free matrices,linear convex combination method and the new criterion,the delay-dependent is given to reach the global synchronization for the considered neural networks.At last,an example is presented to account for the validity of the results.Chapter 3 considers the stochastic stability for stochastic neural networks of neutral type with flexible terminals method and semi-Markov jump parameters.Based on the flexible terminal method and semi-Markov property,designs the corresponding Lyapunov-Krasovskii functional.By using the Jensen integral inequality,the Wirtinger-based integral inequality,reciprocal convex combination technique,and the promoted double-integral inequality to estimate the derivative of the Lyapunov-Krasovskii functional.Finally,gives the stability criterion.