Closed-loop fault diagnosis based on a nonlinear process model and automatic fuzzy rule generation

A complete procedure for leak detection and diagnosis in a complex heat exchanger using data-driven fuzzy models

“…(5) describes a linear parameter-varying (LPV) model which is generally considered grey-box as it possesses a partly transparent structure [9]. This class of models is well suited for generation of symptoms which can be used for fault detection and diagnosis (FDD) purposes.…”

“…This approach should reduce the effects of uncertainties on the FDD system and enhance its sensitivity to faults. Developing such models from physical laws and/or measurements is still of substantial interest [7][8][9][10][11]. An efficient method to obtain the required nonlinear dynamic models is to resort to models based on fuzzy logic reasoning.…”

Further potentials of CAQ tools for fault handling in automotive manufacturing processes

“…More recently, the potential of artificial intelligence in process monitoring was investigated. Successful applications have been reported in the literature using neural networks [6,7], fuzzy logic and hybrid techniques [8].…”

Fault diagnosis in chemical processes and equipment with feedbacks

“…Among possible approaches to on-line diagnostics of faults in the objects under consideration, the most promising ones are fuzzy DM Chemometrics and Intelligent Laboratory Systems 126 (2013) 123-128 [10,11] and observer-based models [1,12]; here Kalman filters are often used. In the latter case, the object DM is formed in a state space.…”