Finite-time memory fault detection filter design for nonlinear discrete systems with deception attacks

Abstract

In this paper, the finite-time memory fault detection filter (MFDF) is designed for nonlinear discrete systems with randomly occurring deception attacks, where the phenomenon of the randomly occurring deception attacks is characterised by a Bernoulli distributed random variable with known probability. To be specific, the finite-horizon data are employed to construct the MFDF. The main purpose of this paper is to design the MFDF such that, for all nonlinearities, external disturbances and randomly occurring deception attacks, the resultant augmented system is finite-time stable and attains the H ∞ performance. Accordingly, some sufficient conditions are developed to guarantee the existence of the desired fault detection filter parameters, where the solvability of the addressed problem is verified by the feasibility of certain matrix inequalities. Moreover, in order to show that the MFDF has better detection performance, a non-memory fault detection filter (NMFDF) is constructed to compare with the MFDF. Finally, two numerical simulations are utilised to illustrate the effectiveness of the proposed fault detection strategies and explain the superiority of the proposed MFDF.

A novel PID-like particle swarm optimizer: on terminal convergence analysis

“…Experimental results have exhibited the superiorities of designed PIDLPSO algorithm over other state-of-the-art PSO variants on four wide-ranging benchmark functions including both one-peak and multi-peak cases. In the future, we will research into some new directions which include, but are not limited to, the investigations on (1) how to analyze the convergence of the modified PSO algorithms with timevarying parameters [55] and (2) how to apply the PIDLPSO algorithm to other research fields such as deep learning [10,15,26,38,43,54,58], fault detection [3,11,12], signal processing [23,25,27,34,35,37,40,45,47,51] and multi-objective optimization [9,53].…”

Empower parameterized generative adversarial networks using a novel particle swarm optimizer: algorithms and applications

“…The real edge of the crack other GAN variants (e.g. the Wasserstein GAN, the least squares GAN) and machine learning algorithms for crack detection on NDT data [2,31,32]; (2) improve the search ability of the neighbourhood-PSO algorithm by designing a novel strategy to assign appropriate weights to each individual particle in the neighbourhood [1,13,25,26,46]; (3) design an adaptive control strategy to developed GAN-based crack detection algorithm [4, 14, 18, 27-29, 33, 39, 49, 52, 53, 58]; (4) study the stability of the ECPT system [6,7,15,17,19,20,36,38,50,56,57]; and (5) apply the proposed algorithm to other applications, such as thermal-to-visible image translation and image super-resolution [8,12].…”

Incipient fault prediction for nonlinear stochastic distribution systems

“…Observers are designed for fault detection in stochastic systems in References 26‐28. Finite‐time memory fault detection filters (MFDFs) are designed in Reference 29 for nonlinear discrete systems where deception attacks occur randomly. A deterministic learning‐based method for incipient fault detection in nonlinear dynamic systems is proposed in Reference 30.…”