1. Accuracy is not enough: optimizing for a fault detection delayMatej Šprogar, Domen Verber, 2023, izvirni znanstveni članek Opis: This paper assesses the fault-detection capabilities of modern deep-learning models. It highlights that a naive deep-learning approach optimized for accuracy is unsuitable for learning fault-detection models from time-series data. Consequently, out-of-the-box deep-learning strategies may yield impressive accuracy results but are ill-equipped for real-world applications. The paper introduces a methodology for estimating fault-detection delays when no oracle information on fault occurrence time is available. Moreover, the paper presents a straightforward approach to implicitly achieve the objective of minimizing fault-detection delays. This approach involves using pseudo-multi-objective deep optimization with data windowing, which enables the utilization of standard deep-learning methods for fault detection and expanding their applicability. However, it does introduce an additional hyperparameter that needs careful tuning. The paper employs the Tennessee Eastman Process dataset as a case study to demonstrate its findings. The results effectively highlight the limitations of standard loss functions and emphasize the importance of incorporating fault-detection delays in evaluating and reporting performance. In our study, the pseudo-multi-objective optimization could reach a fault-detection accuracy of 95% in just a fifth of the time it takes the best naive approach to do so. Ključne besede: artificial neural networks, deep learning, fault detection, accuracy, multi-objective optimization Objavljeno v DKUM: 30.11.2023; Ogledov: 363; Prenosov: 27
Celotno besedilo (478,93 KB) Gradivo ima več datotek! Več... |
2. Fault detection of an industrial heat-exchanger : a model-based approachDejan Dragan, 2011, izvirni znanstveni članek Opis: One of the key issues in modelling for fault detection is how to accommodate the level of detail of the model description in order to suit the diagnostic requirements. The paper addresses a two-stage modelling concept to an industrial heat exchanger, which is located in a tyre factory. Modelling relies on combination of prior knowledge and recorded data. During the identification procedure, the estimates of continuous model parameters are calculated by the least squares method and the state variable filters (SVF). It is shown that the estimates are largely invariant of the bandwidth of the SVFs. This greatly reduces the overall modelling effort and makes the whole concept applicable even to less experienced users. The main issues of the modelling procedure are stressed. Based on the process model a simple detection system is derived. An excerpt of the results obtained on operating records is given. Ključne besede: industrijski prenosniki toplote, zaznavanje napak, nadzor procesov, odkrivanje napak na osnovi modela, modeliranje, identifikacija, industrial heat exchanger, fault detection, condition monitoring, model-based detection, modelling, identification Objavljeno v DKUM: 10.07.2015; Ogledov: 1878; Prenosov: 30
Povezava na celotno besedilo |
3. Crack identification in gear tooth root using adaptive analysisAleš Belšak, Jože Flašker, 2007, izvirni znanstveni članek Opis: Problems concerning gear unit operation can result from various typical damages and faults. A crack in the tooth root, which often leads to failure in gear unit operation, is the most undesirable damage caused to gear units. This article deals with fault analyses of gear units with real damages. A laboratory test plant has been prepared. It has been possible to identify certain damages by monitoring vibrations. In concern to a fatigue crack in the tooth root significant changes in tooth stiffness are more expressed. When other faults are present, other dynamic parameters prevail. Signal analysis has been performed also in concern to a non-stationary signal, using the adaptive transformation to signal analysis. Ključne besede: machine elements, gears, fatigue crack, fault detection, vibrations, adaptive signal analysis, engineering diagnostics Objavljeno v DKUM: 31.05.2012; Ogledov: 2206; Prenosov: 81
Povezava na celotno besedilo |