Visualizing Uncertainty in Reliability Functions with Application to Aero Engine Overhaul

Ebden, Mark; Stranjak, Armin; Roberts, Stephen A.

doi:10.1111/j.1467-9876.2009.00692.x

Cited by 7 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A finite mixture model is used to capture the combination of individual performances. Ebden, Stranjak, and Roberts (2010) also propose a finite mixture model to describe jet engine failure modes. Here, the mixed Weibull distribution is estimated from a large data set of around 300 jet engines.…”

Section: Theory and Related Workmentioning

confidence: 99%

A Case for the Use of Data-driven Methods in Gas Turbine Prognostics

et al. 2017

View full text Add to dashboard Cite

The goal of data-driven methods is to remove dependence on classical models of structured expert judgment and draw insights to causal relationships directly from the data. This paper investigates the potential of using data-driven methods, namely uni-variate multiple linear regression, k-nearest neighbors, feed-forward neural networks, random forests and linear support vector regression to predict the end of life(EOL) and remaining useful life (RUL) of engineering systems. The algorithms are demonstrated on a real-world largescale dataset consisting of a multidimensional time series of health monitoring indicators collected from a set of commercial aircraft gas turbine engines. A stratified version of10-fold cross-validation is used to compare the prognostics performance of the five prognostics models. An experiencebased Weibull model is chosen as the baseline method. Models are evaluated according to established metrics in the field including median absolute error, median absolute deviation and relative accuracy. The prediction results indicate that support vector regression and random forests are the most accuratemodels. Neural networks and k-nearest neighbors also show improved forecast skill compared to the baseline model while beating the more traditional technique of linear regression. In regards to error spread, results are not as expressive even though all the selected data-driven methods provide good results, outperforming the baseline.

show abstract

Section: Theory and Related Workmentioning

confidence: 99%

A Case for the Use of Data-driven Methods in Gas Turbine Prognostics

et al. 2017

View full text Add to dashboard Cite

show abstract

“…The authors found that such model could help not only to describe the engine reliability but also to investigate the interdependent effects among the disruption modes. A finite mixture model is also used to describe jet engine failure modes in [10]. The mixed Weibull distribution is estimated from a large data set comprised of 325 jet engines.…”

Section: Introductionmentioning

confidence: 99%

Aircraft on-condition reliability assessment based on data-intensive analytics

Baptista

Medeiros

Malere

et al. 2017

2017 IEEE Aerospace Conference

View full text Add to dashboard Cite

The implementation of condition-based maintenance continues to face several challenges especially in the aeronautics field. While it is true that time-based maintenance dominates the industry today, it is believed that condition monitoring could yield promising results with a better compromise on cost over effectiveness in the long run. The aim of condition-based monitoring in aeronautics is, based on the available system data (e.g., flight, event and maintenance data), to evaluate the current health state of an aircraft component and to estimate its remaining useful life. Several approaches have been studied in condition-based maintenance with the most promising being data-driven modeling. This paper proposes a comparison of a set of data-driven modeling techniques to perform prognostics on a critical component of the jet engine bleed system. The novelty of our work is twofold. First, we perform this comparative study on a real case study of a critical valve of the aircraft bleed system. Fielded data from different data sources are used in the models. To our knowledge, this is the first case study that merges data from the computer central maintenance system (fault messages), maintenance data, and flight data on a prognostics system. Second, a variety of data-driven techniques are compared from neural nets to regression support machines. The models are compared using the standard metrics of absolute, mean, and squared errors. A regressive accuracy curve is also used to compare the models along different prediction window sizes. The results show the best model comprised information from all data sources. The data that most contributed to the performance improvement was the maintenance, flight and fault data, in this order. This result comes to reinforce the notion that it is more difficult to extract quantitative information from fault events than flight data with data-driven regressive methods.

show abstract

“…These are the limit distributions for the maxima of exponential, heavy-tailed, and light-tailed distributions, respectively. The Weibull is that distribution typically used to estimate the distribution of component lifetimes in reliability theory [ 10 , 21 , 32 ].…”

Section: Introduction To Extreme Value Theorymentioning

confidence: 99%

Extending the Generalised Pareto Distribution for Novelty Detection in High-Dimensional Spaces

Clifton

Hugueny

et al. 2013

J Sign Process Syst

View full text Add to dashboard Cite

Novelty detection involves the construction of a “model of normality”, and then classifies test data as being either “normal” or “abnormal” with respect to that model. For this reason, it is often termed one-class classification. The approach is suitable for cases in which examples of “normal” behaviour are commonly available, but in which cases of “abnormal” data are comparatively rare. When performing novelty detection, we are typically most interested in the tails of the normal model, because it is in these tails that a decision boundary between “normal” and “abnormal” areas of data space usually lies. Extreme value statistics provides an appropriate theoretical framework for modelling the tails of univariate (or low-dimensional) distributions, using the generalised Pareto distribution (GPD), which can be demonstrated to be the limiting distribution for data occurring within the tails of most practically-encountered probability distributions. This paper provides an extension of the GPD, allowing the modelling of probability distributions of arbitrarily high dimension, such as occurs when using complex, multimodel, multivariate distributions for performing novelty detection in most real-life cases. We demonstrate our extension to the GPD using examples from patient physiological monitoring, in which we have acquired data from hospital patients in large clinical studies of high-acuity wards, and in which we wish to determine “abnormal” patient data, such that early warning of patient physiological deterioration may be provided.

show abstract

Visualizing Uncertainty in Reliability Functions with Application to Aero Engine Overhaul

Cited by 7 publications

References 8 publications

A Case for the Use of Data-driven Methods in Gas Turbine Prognostics

A Case for the Use of Data-driven Methods in Gas Turbine Prognostics

Aircraft on-condition reliability assessment based on data-intensive analytics

Extending the Generalised Pareto Distribution for Novelty Detection in High-Dimensional Spaces

Contact Info

Product

Resources

About