New alterable terminal method on H∞ State Estimation of Delayed Static Neural Networks

This study is concerned with the issue of H
_∞
state estimation for static neural networks (SNNs) with time-varying delays. Firstly, we employ a new alterable terminal method (ATM), which is proposed by applying convex combination method and introducing a tunable parameter. Secondly, on the basis of the ATM, we choose an augmented Lyapunov-Krasovskii functional (LKF), where the integral interval connected with time-delay variables is unfixed. Thirdly, we introduce the convex combination approach to represent the time-varying delay derivative h(t). By using reciprocally convex approach and Wirtinger-based integral inequality, a novel condition is obtained so that the error system is globally asymptotically stable with a guaranteed H
_∞
performance index γ. Fourthly, better performance index and the desired control action K are achieved under the above methods. Finally, a numerical example is showed to illustrate the effectiveness and superiority of the developed methods.