Parametric uncertainty in sensor fault detection for a turbotan jet engine

Abstract

Abstract-A method for detecting sensor faults in a turbofan engine is presented. The pmpnsed method consists of an observer with integral action and an adaptive detection threshold. The threshold is computed with the assumption of parametric uncertainty in the process model. Successful simulations with sensnr data from an RMlZ jet engine shows that the method is capable of detecting even a very smaU increase in sensor noise promptly without generating false alarms.

Robust time-varying thresholds for supervision of valves in a flotation process

“…In [15], robustness against unstructured additive uncertainty in the frequency domain is considered, [16] presents a robust threshold for detection of clogging in the coal injection lines of a blast furnace and [17] for sensor fault detection for a turbofan jet engine. Other recent publications of robust thresholds are [18] and [19].…”

“…by windowing or exponential forgetting, in [17] the exponential forgetting weighting function µe −µt is used. Note also that the evaluation signal (14) proposed in [23] can be expressed as s = S Θ d e, where Θ d is the dirac delta function.…”

An Interval Intelligent-based Approach for Fault Detection and Modelling

“…In fact, not only uncertainties exist in the analytical model A. Khosravi, J. Armengol and E. R. Gelso are with the Institut d'Informatica i Aplicacions, Universitat de Girona, E-17071 Girona, Catalonia, Spain {khosravi,armengol,ergelso}@eia.udg.es of the system owing some complicacy in modeling, but also they influentially mischievously impress measurements. A countless number of researchers have devoted a great deal of effort and time in last decades to deal with these uncertainties and confine their negative effects on fault detection and false alarm rates [5], [6], [7], [8], [9], [10].…”

Linear Optimization of Parameters in Dynamic Threshold Generators

“…A process where fault detection algorithms may be a significant advantage is the jet engine in a single engine aircraft where faults can have catastrophic consequences. A fault detection algorithm with a dynamic detection threshold for a sensor in a turbofan engine is presented in (Johansson and Norlander 2003). There, constant parameter uncertainties are assumed and the uncertainty bound design parameters are tuned manually.…”

“…There, constant parameter uncertainties are assumed and the uncertainty bound design parameters are tuned manually. In (Johansson and Bask 2005) the approach from (Johansson and Norlander 2003) is generalized to allow time-varying parameter uncertainties but the bounds are still tuned manually. In this paper, an automatic method to determine the threshold design parameters, substituting the upper bounds, is derived.…”