Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
On-board Clutch Slippage Detection and Diagnosis in Heavy Duty Machine
Luleå University of Technology, Department of Engineering Sciences and Mathematics, Product and Production Development. Volvo Construction Equipment, Eskilstuna, Sweden.ORCID iD: 0000-0003-1691-4387
RISE SICS Västerås, Sweden.
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.ORCID iD: 0000-0002-2356-7830
Department of Applied Signal Processing, Blekinge Institute of Technology, Karlskrona, Sweden.
Show others and affiliations
2018 (English)In: International Journal of Prognostics and Health Management, ISSN 2153-2648, E-ISSN 2153-2648, Vol. 9, no 1, article id 007Article in journal (Refereed) Published
Abstract [en]

In order to reduce unnecessary stops and expensive downtime originating from clutch failure of construction equipment machines; adequate real time sensor data measured on the machinein combination with feature extraction and classification methods may be utilized.

This paper, based on a study at Volvo Construction Equipment,presents a framework with feature extraction methods and an anomaly detection module combined with Case-Based Reasoning (CBR) for on-board clutch slippage detection and diagnosis in a heavy duty equipment. The feature extraction methods used are Moving Average Square Value Filtering (MASVF) and a measure of the fourth order statistical properties of the signals implemented as continuous queries over data streams. The anomaly detection module has two components,the Gaussian Mixture Model (GMM) and the Logistics Regression classifier. CBR is a learning approach that classifies faults by creating a new solution for a new fault case from the solution of the previous fault cases. Through use of a data stream management system and continuous queries (CQs), the anomaly detection module continuously waits for a clutch slippage event detected by the feature extraction methods, the query returns a set of features which activates the anomaly detection module. The first component of the anomaly detection module trains a GMM to extracted features while the second component uses a Logistic Regression classifier for classifying normal and anomalous data. When an anomalyis detected, the Case-Based diagnosis module is activated for fault severity estimation.

Place, publisher, year, edition, pages
PHM Society , 2018. Vol. 9, no 1, article id 007
National Category
Engineering and Technology Control Engineering Other Mechanical Engineering
Research subject
Computer Aided Design; Control Engineering
Identifiers
URN: urn:nbn:se:ltu:diva-67976Scopus ID: 2-s2.0-85044281699OAI: oai:DiVA.org:ltu-67976DiVA, id: diva2:1191297
Note

Validerad;2018;Nivå 1;2018-03-19 (rokbeg)

Available from: 2018-03-17 Created: 2018-03-17 Last updated: 2018-05-16Bibliographically approved
In thesis
1. Data Driven Condition Monitoring for Transmission and Axles
Open this publication in new window or tab >>Data Driven Condition Monitoring for Transmission and Axles
2018 (English)Doctoral thesis, comprehensive summary (Other academic)
Alternative title[sv]
Datadriven övervakning för transmission och axlar
Abstract [en]

As the requirements to improve up-time and thus to reduce costly down-time con-tinuously increases, the construction equipment business focuses on more and newways to increase ability and sensitivity of early fault detection of critical compo-nents and parts in order to prevent failure. Failure of critical components in theheavy duty machine may lead to unnecessary stops and expensive downtime. Withmore features added to the heavy duty construction equipment, its complexity in-creases and early fault detection of certain components becomes more challengingdue to too many fault codes generated when a failure occurs. Hence, the need tocomplement the present onboard diagnostic methods with more sophisticated diag-nostic methods for adequate condition monitoring of the heavy duty constructionequipment in order to improve uptime. Further, reduced downtime leads to im-proved customer satisfaction, reduced warranty and service cost. In addition, thisupgrade result in the construction equipment business staying competitive with im-provement in sales and profit.

Heavy duty construction equipment is often equipped with a driveline whichconsists of major components, such as torque converter, gearbox, clutches, bearingsand axles. The driveline enables the transferring of torque from the engine to thegearbox, with the clutches enabling automatic gear ratio changes, and this drivingtorque from the gearbox is further transmitted to the wheels via the axles. Thesemajor components of a driveline may be considered as crucial components whosefailure may result in costly downtime. Since the current on-board diagnostic sys-tems use simple rules and maps to carry out diagnosis, most failures are not easy todiagnose as a result of too many fault codes being generated when there is a fail-ure. This means that, the engineers and technicians may have to spend substantialamount of time to identify the failure and root-cause. As a result, where major driv-eline parts are involved, this may cause the machine to stand still until the problemis identified and repaired, with a negative impact on customer satisfaction.

In this thesis, condition monitoring methods are presented with the purpose toprovide a diagnostic framework possible to implement onboard for monitoring ofcritical driveline parts in order to improve uptime.

In this thesis the gap in condition monitoring of major driveline components in an actual machine is addressed. A methodology for monitoring the health of theautomatic transmission and axles onboard the machine using vibration signals andavailable CAN-bus signals has been developed. Furthermore, this thesis presentsa vibration based diagnostic framework for the monitoring of the torque converter,gearbox, bearings and axles. For the development of this diagnostic framework,sensor data from the gearbox, torque converter, bearings and axles are considered.Further, the feature extraction of the data collected has been carried out using or-der analysis technique and adequate signal processing methods, which includes,Adaptive Line Enhancer, Order Power Spectrum and Order Modulation Spectrumrespectively. In addition, Bayesian learning was utilized for learning of the extractedfeatures onboard. The results indicate that the vibration properties of the gearbox,torque converter, bearings and axle are relevant for early fault detection of the driv-eline. Furthermore, vibration provides information about the internal features ofthese components for detecting deviations from normal behavior.

A different approach was utilized for the monitoring of the automatic transmis-sion clutches. The feature extraction methods utilized for the monitoring of theautomatic transmission clutches are based on moving average square value filter-ing and a measure of the fourth order statistical properties of the CAN-bus signals.Results show that the feature extraction methods provide an indication of clutchslippage deviations. This thesis also includes an investigation of clutch slippage de-tection from driveline vibrations based on spectrogram and spectral Kurtosis meth-ods.

In this way, the developed methods may be implemented onboard for the con-tinuous monitoring of these critical driveline parts of the heavy duty constructionequipment so that if their health starts to degrade a service and/or repair may bescheduled well in advance of a potential axle failure and in that way the downtimeof a machine may be reduced and costly replacements and repairs avoided.

Place, publisher, year, edition, pages
Luleå: Luleå University of Technology, 2018. p. 85
Series
Doctoral thesis / Luleå University of Technology 1 jan 1997 → …, ISSN 1402-1544
National Category
Other Mechanical Engineering
Research subject
Computer Aided Design
Identifiers
urn:nbn:se:ltu:diva-68180 (URN)978-91-7790-090-0 (ISBN)978-91-7790-091-7 (ISBN)
Public defence
2018-05-21, E632, Luleå, 10:00 (English)
Opponent
Supervisors
Available from: 2018-04-06 Created: 2018-04-05 Last updated: 2018-05-24Bibliographically approved

Open Access in DiVA

fulltext(539 kB)23 downloads
File information
File name FULLTEXT01.pdfFile size 539 kBChecksum SHA-512
f29cdb976a0964847082e3f08f26d3a8b9fa8e42fb54549dbcc9fc706b15862f24cc6125b1bf8a836cc55ed8aa389d5f3b50012832da1f6bc2aa6191e7f713d9
Type fulltextMimetype application/pdf

Other links

Scopushttp://www.phmsociety.org/references/ijphm-archives/2018

Authority records BETA

Källström, ElisabethLindström, John

Search in DiVA

By author/editor
Källström, ElisabethLindström, John
By organisation
Product and Production DevelopmentSignals and Systems
In the same journal
International Journal of Prognostics and Health Management
Engineering and TechnologyControl EngineeringOther Mechanical Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 23 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 73 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf