Predictive maintenance demonstrator dataset with individual load histories

Documentation of the data
datacite.description.TechnicalInfo

A thorough documentation of this dataset, including experimental setup, creation of the signals, postprocessing and data access is provided in the textfile dataset_description.txt in the dataset.

Countries to which the data refer
datacite.geolocation.iso3166

GERMANY

References to related material
datacite.relatedItem.References

http://dx.doi.org/10.25532/OPARA-189

Description of the data
datacite.resourceType

The research field of predictive maintenance aims to develop methods that are capable of predicting component failure before it occurs. This way, maintenance can be conducted with the component is still intact, while at the same time increasing resource efficiency by exploiting the fatigue strength of the component. In order to enable predictive maintenance, sensor data of the component is required, from which the predictions can be generated. Furthermore, the individual loads acting on each component of a system depend on the operating conditions of the systems and are therefore different between different systems. Failure therefore occurs as a result of individual load histories. Virtual sensing methods aim to predict unmeasured physical quantities from available measurement data. These methods offer significant benefits to predictive maintenance, since virtual sensors can be used to estimate quantities that are difficult to measure. In fleet situations, where many identical systems with different usage histories are monitored, virtual sensors can also be deployed as a replacement for some sensors of the system. This can be done in a data-driven manner by installing sensors for all quantities of interest at one reference system, whose measurement data is used to parameterize the virtual sensor model, so that it can then be deployed for the remaining standard systems. In many real applications, the time to failure is in the range of years, complicating the development and validation of predictive maintenance and virtual sensing approaches. The aim of this dataset is therefore to provide a demonstrator example where failure occurs based on individual load histories. Its sensor setup is designed to provide a virtual sensor use case with independent training and testing data, so that the dataset can be used for algorithm development and benchmarking purposes.

Type of the data
datacite.resourceTypeGeneral

Dataset

Total size of the dataset
datacite.size

2239844090

Author
dc.contributor.author

Heindel, Leonhard

Author
dc.contributor.author

Hantschke, Peter

Author
dc.contributor.author

Kästner, Markus

Upload date
dc.date.accessioned

2025-04-14T15:27:54Z

Publication date
dc.date.available

2025-04-14T15:27:54Z

Data of data creation
dc.date.created

2025-04-04

Publication date
dc.date.issued

2025-04-14

Abstract of the dataset
dc.description.abstract

Predictive maintenance aims to develop methods that are capable of predicting component failure before it occurs. Virtual sensing methods predict unmeasured physical quantities from available measurement data. These methods offer significant benefits to predictive maintenance, since virtual sensors can be used to estimate quantities that are difficult to measure. In many real applications, the time to failure is in the range of years, complicating the development and validation of predictive maintenance and virtual sensing approaches. This dataset provides a demonstrator example where failure occurs based on individual load histories. The sensor setup consists of simple notched steel specimens, which are clamped between two servo-hydraulic cylinders of a fatigue test bench. It is designed to provide a virtual sensor use case with independent training and testing data, so that the dataset can be used for algorithm development and benchmarking purposes.

Public reference to this page
dc.identifier.uri

https://opara.zih.tu-dresden.de/handle/123456789/1427

Public reference to this page
dc.identifier.uri

https://doi.org/10.25532/OPARA-812

Publisher
dc.publisher

Technische Universität Dresden

Licence
dc.rights

Attribution 4.0 Internationalen

URI of the licence text
dc.rights.uri

http://creativecommons.org/licenses/by/4.0/

Specification of the discipline(s)
dc.subject.classification

4

Title of the dataset
dc.title

Predictive maintenance demonstrator dataset with individual load histories

Research instruments
opara.descriptionInstrument

servo-hydraulic fatigue test bench

Research instruments
opara.descriptionInstrument

acceleration sensors

Research instruments
opara.descriptionInstrument

force sensors

Research instruments
opara.descriptionInstrument

displacement sensors

opara.descriptionObject
opara.descriptionObject

virtual sensing

opara.descriptionObject
opara.descriptionObject

fatigue

Underlying research object
opara.descriptionObject.NonPhysicalObject

predictive maintenance

Software
opara.descriptionSoftware.ResourceViewing

Python

Project abstract
opara.project.description

The aim of the project is to increase the availability of a tramcar fleet by evaluating standard sensors that are available in every tramcar. Insights into the condition of tramcar and track enable target-oriented fleet usage and reduce follow-up maintenance costs. The increase in rail reliability increases the acceptance of public transport among the population. Financial advantages arise for operators and users.

Public project website(s)
opara.project.publicReference

https://tu-dresden.de/ing/maschinenwesen/ifkm/nefm/forschung/projekte/LRVTwin

Project title
opara.project.title

LRVTwin - A digital twin for light rail vehicles
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Heindel_2025_PM_demonstrator.zip
Size:
2.09 GB
Format:
Unknown data format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
4.66 KB
Format:
Item-specific license agreed to upon submission
Description:
Attribution 4.0 International