Intelligent Fault Diagnosis and Prognosis for Engineering Systems

Vachtsevanos, George; Lewis, Frank L.; Roemer, Michael; Hess, Andrew; Wu, Biqing

doi:10.1002/9780470117842

Cited by 850 publications

(637 citation statements)

References 31 publications

Supporting

Mentioning

625

Contrasting

Unclassified

Order By: Relevance

“…If one catches the fault at 4 percent severity, one needs replace only the component. If the fault is not caught until 10 percent severity, the subsystem must be replaced, and at failure, the entire system must be replaced [6]. Thus, predictions about fault severity and impending failures are essential.…”

Section: Diagnostics Stepmentioning

confidence: 99%

Learning in dynamic environments: application to the diagnosis of evolving systems

Mouchaweh¹,

Ayad²,

Malki³

2011

Proceedings of the 7th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT-2011)

View full text Add to dashboard Cite

In dynamic environments, data characteristics may drift over time. This leads to deteriorate dramatically the performance of incremental learning algorithms over time. This is because of the use of data which is no more consistent with the characteristics of new incoming one. In this paper, an approach for learning in dynamic environments is proposed. This approach integrates a mechanism to use only the recent and useful patterns to update the classifier without a "catastrophic forgetting". This approach is used for the acoustic leak detection in the steam generator unit of the nuclear power generator "Prototype Fast React".

show abstract

Section: Diagnostics Stepmentioning

confidence: 99%

Learning in dynamic environments: application to the diagnosis of evolving systems

Mouchaweh¹,

Ayad²,

Malki³

2011

Proceedings of the 7th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT-2011)

View full text Add to dashboard Cite

show abstract

“…The new technology frontier is clearly prognosis which adds the predictive element to what has gone before 5,6 . This new technology seeks to be able to identify problems before they happen so that corrective action can be taken prior to loss of service.…”

Section: Sensor-based Diagnosis and Prognosismentioning

confidence: 99%

Integrated Vehicle Health Management in the Automotive Industry

Holland¹

2010

Health Management

View full text Add to dashboard Cite

“…Using the prediction of the degradation paths and the defined fault tolerance limits, remaining useful life (RUL) can be estimated [56]. Predicting the remaining useful life based on the degradation prediction is an essential element of the health management of complex systems [56].…”

Section: Introductionmentioning

confidence: 99%

“…For predicting dynamic degradations several approaches have been proposed. Physical models and filtering approaches, such as Kalman filters [56] and Particle filtering [21], [28] belong to state-of-the-art methods applied in degradation modelling. There are some similarities between time series regression and filtering.…”

Section: Introductionmentioning

confidence: 99%

Predicting component reliability and level of degradation with complex-valued neural networks

Fink

Zio

Weidmann

2014

Reliability Engineering & System Safety

104

View full text Add to dashboard Cite

. Predicting component reliability and level of degradation with complex-valued neural networks. Reliability Engineering and System Safety, Elsevier, 2014, 121, pp.198-206 AbstractIn this paper, multilayer feedforward neural networks based on multi-valued neurons (MLMVN), a specific type of complex valued neural networks, are proposed to be applied to reliability and degradation prediction problems, formulated as time series. MLMVN have demonstrated their ability to extract complex dynamic patterns from time series data for mid-and long-term predictions in several applications and benchmark studies. To the authors' knowledge, it is the first time that MLMVN are applied for reliability and degradation prediction. MLMVN is applied to a case study of predicting the level of degradation of railway track turnouts using real data. The performance of the algorithms is first evaluated using benchmark study data. The results obtained in the reliability prediction study of the benchmark data show that MLMVN outperforms other machine learning algorithms in terms of prediction precision and is also able to perform multi-step ahead predictions, as opposed to the previously best performing benchmark studies which only performed up to two-step ahead predictions. For the railway turnout application, MLMVN confirm the good performance in the long-term prediction of degradation and does not show accumulating errors for multi-step ahead predictions.

show abstract

Intelligent Fault Diagnosis and Prognosis for Engineering Systems

Cited by 850 publications

References 31 publications

Learning in dynamic environments: application to the diagnosis of evolving systems

Learning in dynamic environments: application to the diagnosis of evolving systems

Integrated Vehicle Health Management in the Automotive Industry

Predicting component reliability and level of degradation with complex-valued neural networks

Contact Info

Product

Resources

About