Synthetic data generation in hybrid modelling of rolling element bearings

Leturiondo, Urko; Mishra, Madhav; Galar, Diego; Salgado, Óscar

doi:10.1784/insi.2015.57.7.395

Cited by 15 publications

(11 citation statements)

References 10 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This occurs because maintainers replace critical elements in early stages of degradation for safety, reliability, economic, and environmental reasons. Therefore, it is challenging to acquire data in faulty stages of the system and advanced stages of degradation [3].…”

Section: Problem Discussionmentioning

confidence: 99%

Hybrid Models and Digital Twins for Condition Monitoring: HVAC System for Railway

Gálvez¹,

Rubio²,

Seneviratne³

et al. 2021

SNE

View full text Add to dashboard Cite

Safety passenger transportation is more important than efficiency or reliability. Therefore, it is vital to maintain the proper condition of the equipment related to the passengers' comfort and safety. This manuscript presents the methodology of complete development and implementation of both hybrid model and digital twin 3.0 for an HVAC in railways. The objective of this is to monitor the condition of the HVAC where it matters to the comfort and safety of the passengers in the trains. The level 3.0 of digital twin will be developed for the diagnosis and prognosis of HVAC by using hybrid modeling. The description illustrated in this paper is focused on the methodology used to implement a hybrid model-based approach, and both the need and advantages of using hybrid model approaches instead of data-based approaches. The development considers the importance of safety and environmental risks, which are included in the risk quantification of failure modes. Railway's maintainers replace critical components in early stages of degradation; thus, the use of a data-driven model loses essential information related to advanced stages of degradation which might decrease the accuracy of the maintenance instructions provided. Physics-based model can be used to generate synthetic data to overcome the lack of data in advanced stages of degradation, and then, the synthetic data can be combined with the real data, which is collected by sensor located in the real system, to build the data-driven model. The combination leads to form hybrid-model based approach with a large number of failure modes that were unpredictable. Finally, the outcome is beneficial for the proper functioning of systems; hence, safety of the passengers.

show abstract

Section: Problem Discussionmentioning

confidence: 99%

Hybrid Models and Digital Twins for Condition Monitoring: HVAC System for Railway

Gálvez¹,

Rubio²,

Seneviratne³

et al. 2021

SNE

View full text Add to dashboard Cite

show abstract

“…In this figure, a plot of RMS versus the kurtosis of data series for each track segment is built. RMS and kurtosis are widely-used statistical indicators used in signal characteristic analysis 20 . The resulting clusters seen in Fig.…”

Section: Datamentioning

confidence: 99%

Dealing with disruptions in railway track inspection using risk-based machine learning

Kaewunruen

Osman

2023

Sci Rep

View full text Add to dashboard Cite

Unplanned track inspections can be a direct consequence of any disruption to the operation of on-board track geometry monitoring activities. A novel response strategy to enhance the value of the information for supplementary track measurements is thus established to construct a data generation model. In this model, artificial (synthetic) data is assigned on each measurement point along the affected track segment over a short period of time. To effectively generate artificial track measurement data, this study proposes a NARX (nonlinear autoregressive with exogenous variables) model, which incorporates short-range memory dependencies in the dependent variable and integrates interdependent effects from external factors. Nonlinearities in the proposed model have been determined using an artificial neural network that allowed fast computation of a mapping function in line with the needs of effective disruption management. The risk of over fitting the data generation model, which reflected its generalisation ability, has been effectively managed through risk aversion concept. For the model evaluation, the deviation of track longitudinal level has been taken as a case study, predicted using its degradation rate and track alignment and gauge as exogenous variables. Simulation results on two datasets that are statistically different showed that the data generation model for disrupted track measurements is reliable, accurate, and easy-to-use. This novel model is an essential breakthrough in railway track integrity prediction and resilient operation management.

show abstract

“…Traditional stationary signal processing techniques cannot be applied effectively in a time-varying conditions context [10]. Due to the time-varying conditions both for the operating conditions and the healthy or faulty state, can be necessary to properly generate data related to the operating context of the analyzed system [11]. The role of the domain experts in the anomaly detection context is crucial even for the definition of the boundaries between normal and anomalous behavior as well as for the economic analysis of the anomalies impact [12] and the extraction of useful details from the raw signals [13].…”

Section: Literature Reviewmentioning

confidence: 99%

Machine learning for anomaly detection and process phase classification to improve safety and maintenance activities

Quatrini

Costantino

Gravio

et al. 2020

Journal of Manufacturing Systems

View full text Add to dashboard Cite

Anomaly detection is a crucial aspect for both safety and efficiency of modern process industries. This paper proposes a two-steps methodology for anomaly detection in industrial processes, adopting machine learning classification algorithms. Starting from a real-time collection of process data, the first step identifies the ongoing process phase, the second step classifies the input data as "Expected", "Warning", or "Critical". The proposed methodology is extremely relevant where machines carry out several operations without the evidence of production phases. In this context, the difficulty of attributing the real-time measurements to a specific production phase affects the success of the condition monitoring. The paper proposes the comparison of the anomaly detection step with and without the process phase identification step, validating its absolute necessity. The methodology applies the decision forests algorithm, as a well-known anomaly detector from industrial data, and decision jungle algorithm, never tested before in industrial applications. A real case study in the pharmaceutical industry validates the proposed anomaly detection methodology, using a 10 months database of 16 process parameters from a granulation process.

show abstract

Synthetic data generation in hybrid modelling of rolling element bearings

Cited by 15 publications

References 10 publications

Hybrid Models and Digital Twins for Condition Monitoring: HVAC System for Railway

Hybrid Models and Digital Twins for Condition Monitoring: HVAC System for Railway

Dealing with disruptions in railway track inspection using risk-based machine learning

Machine learning for anomaly detection and process phase classification to improve safety and maintenance activities

Contact Info

Product

Resources

About