Dissipativity, Fault Detection and Reachable Set Analysis and Synthesis for A Class of Delayed Neural Networks

This paper investigates the analysis and synthesis of neural networks with time delays and Markov jump parameters via the Lyapunov-Krasovskii functional method. Fristly, less-conservative stability criterion is derived for delayed neural networks. Secondly, dissipativity analysis is developed for neural networks with two-delay components. Thirdly, a fault detection filter is designed for delayed Markov jump neural networks. Finally, the problem of reachable set estimation and controller design is solved for delayed Markov jump neural networks with mismatched modes. In conclusion, the thesis gives insightful investigations on analysis and synthesis of neural networks with time-varying delays and Markov jump parameters.