Finite-Time State Estimation of Switched Neural Networks with Both Time-Varying Delays and Leakage Delay

Abstract

In this thesis, we deal with the issues of the finite-time state estimation (FTSE) for a set of switched neural networks (SNNs), in which the hybrid effects of time-varying delays and leakage delay are taken into consideration. Therefore, the model of SNNs under discussion is quite comprehensive and more practical. In the light of an applicable piecewise Lyapunov-Krasovskii (L-K) functional which has double integral terms, some novel sufficient criteria are put forward with the average dwell time (ADT) technique, so that the estimation error system is finite-time boundedness (FTB). It is crucial to notice that the estimation results in our work are time-delay dependent, which depend on the leakage delay as well as the upper bound of the time-varying delays. The results show that the unknown gain matrix of the state estimator is achieved by solving a series of linear matrix inequalities (LMIs), which can be effortlessly tested with the MATLAB Toolbox. Moreover, by combining with free weight matrix method in the proof process, the results we obtained do not require the differentiability of time-varying delays any more, which is less conservative than some existing results. Finally, an example is performed with its numerical simulations to corroborate the efficiency of the theoretical results.