Resilient finite-time fault detection dissipative-based filter for semi-Markovian jump systems with incomplete measurements

摘要

In this article, the problem of resilient finite-time fault detection dissipative-based filter design is investigated for a class discrete-time semi-Markovian jump systems with incomplete measurements, dynamic quantization and time-varying delay. The considered system includes various network-induced imperfections such as missing measurements, transmission delay and dynamic signal quantization. Specifically, all these imperfections are modeled using stochastic variables following Bernoulli random distribution. In addition, the transition probability of the semi-Markov process is time-varying along with the lower and upper bounds of the transition rate. Eventually, a set of sufficient conditions is derived in terms of linear matrix inequalities (LMIs) by assuming proper Lyapunov-Krasovskii functional and using S-procedure lemma such that the augmented filtering error system is stochastically finite-time bounded with prescribed dissipative performance. Finally, the applicability and usefulness of the proposed filter design is demonstrated through a pulse-width-modulation-boost-converter (PWM) model.