Detection of Intermittent Fault for Discrete-Time Systems with Output Dead-Zone: A Variant Tobit Kalman Filtering Approach

Abstract

This paper is concerned with the intermittent fault detection problem for a class of discrete-time linear systems with output dead-zone. Dead-zone phenomenon exists in many real practical systems due to the employment of low-cost commercial off-the-shelf sensors. Two Bernoulli random variables are utilized to model the dead-zone effect and a variant formation of Tobit Kalman filter is brought forward to generate a residual that can indicate the occurrence of an intermittent fault. A numerical example is presented to demonstrate the effectiveness and applicability of the proposed technique. The statistical performance of the technique is illustrated as well.

Protocol-Based Tobit Kalman Filter Under Integral Measurements and Probabilistic Sensor Failures

“…By introducing new definitions (of the measurement expectation, residual as well as variance) and exploiting a local approximation in calculating censoring probabilities, the TKF is capable of formalizing a fully recursive state estimation paradigm to handle the measurement nonlinearity caused by censored observations. Due to its succinct structure and recursive form, much research enthusiasm has recently been attracted towards the TKF and a number of excellent results have been acquired with successful applications in cooperative localization, fault detection, target tracking, and so forth [12], [21], [23]. Within the TKF framework, the state estimation problem has been tackled in [23] with both censored observations and time-correlated multiplicative noises.…”

Tobit Kalman filtering for fractional‐order systems with stochastic nonlinearities under Round‐Robin protocol

“…As such, the Tobit Kalman filter (TKF) has been proposed in Reference 28 to provide a fully recursive state estimation paradigm for handling censored observations. Due to its succinct structure and recursive form, the TKF has immediately been employed in a broad range of application scenarios, for example, the cooperative localization, fault detection and target tracking, see References 29‐33.…”

“…To mitigate impacts from modeling uncertainties, a modified TKF has been designed in Reference 29 with illustrative examples on unmanned aerial vehicle systems. In addition, the fault detection problem has also been tackled in Reference 33 for systems with dead‐zone‐like censoring.…”

Intermittent fault detection for discrete‐ time linear stochastic systems with time delay

“…The synchronous on‐line detection of bank cable IFs was dealt with in [15] for power systems by using the chaotic spread spectrum sequence. By means of the variant Tobit Kalman filtering approach, the IFs detection was discussed in [16] for discrete‐time LSSs with output dead‐zone. In [17], the IFs detection was investigated for multirate systems with quantisation effects and packet dropout by using sampling‐interval‐dependent filters.…”