mdh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Fault Diagnosis of Industrial Machines using Sensor Signals and Case-Based Reasoning
Mälardalen University, School of Innovation, Design and Engineering. (ISS)
2009 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Industrial machines sometimes fail to operate as intended. Such failures can be more or less severe depending on the kind of machine and the circumstances of the failure. E.g. the failure of an industrial robotcan cause a hold-up of an entire assembly line costing the affected company large amounts of money each minute on hold. Research is rapidly moving forward in the area of artificial intelligence providing methods for efficient fault diagnosis of industrial machines. The nature of fault diagnosis of industrial machines lends itself naturally to case-based reasoning. Case-based reasoning is a method in the discipline of artificial intelligence based on the idea of assembling experience from problems and their solutions as ”cases” for reuse in solving future problems. Cases are stored in a case library, available for retrieval and reuse at any time.By collecting sensor data such as acoustic emission and current measurements from a machine and representing this data as the problem part of a case and consequently representing the diagnosed fault as the solution to this problem, a complete series of the events of a machine failure and its diagnosed fault can be stored in a case for future use.

Place, publisher, year, edition, pages
Västerås: Mälardalens högskola , 2009. , 186 p.
Series
Mälardalen University Press Dissertations, ISSN 1651-4238 ; 76
National Category
Computer Science
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:mdh:diva-6539ISBN: 978-91-86135-32-4 (print)OAI: oai:DiVA.org:mdh-6539DiVA: diva2:226846
Public defence
2009-09-18, Pathos, Mälardalens högskola, R-2, Västerås, 13:00 (English)
Opponent
Supervisors
Available from: 2009-07-13 Created: 2009-07-06 Last updated: 2013-12-03Bibliographically approved
List of papers
1. Using Cased-Based Reasoning Domain Knowledge to Train a Back Propagation NeuralNetwork in order to Classify Gear Faults in an Industrial Robot
Open this publication in new window or tab >>Using Cased-Based Reasoning Domain Knowledge to Train a Back Propagation NeuralNetwork in order to Classify Gear Faults in an Industrial Robot
2008 (English)In: 21st International Congress andExhibition. Condition Monitoring and Diagnostic Engineering Management. COMADEM 2008., Prague: Czech Society for Non-Destructive Testing , 2008, 377-384 p.Conference paper, Published paper (Refereed)
Abstract [en]

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained.

Place, publisher, year, edition, pages
Prague: Czech Society for Non-Destructive Testing, 2008
Keyword
Case-Based Reasoning, Neural Network, Sound recordings, Fault classification
National Category
Computer Science
Research subject
Computer Science
Identifiers
urn:nbn:se:mdh:diva-6535 (URN)978-80-254-2276-2 (ISBN)
Projects
Exact
Available from: 2009-07-13 Created: 2009-07-06 Last updated: 2009-07-13Bibliographically approved
2. Agent-Based Monitoring using Case-Based Reasoning for Experience Reuse and Improved Quality
Open this publication in new window or tab >>Agent-Based Monitoring using Case-Based Reasoning for Experience Reuse and Improved Quality
2009 (English)In: Journal of Quality in Maintenance Engineering, ISSN 1355-2511, Vol. 15, no 2, 179-192 p.Article in journal (Refereed) Published
Abstract [en]

Purpose – The purpose with this paper is to propose an agent-based condition monitoringsystem for use in industrial applications. An intelligent maintenance agent is described that isable to autonomously perform necessary actions and/or aid a human in the decision makingprocess. An example is presented as a case-study from manufacturing of industrial robots.Design/methodology/approach – The paper is mainly based on a case-study performed at alarge multi-national company aiming to explore the usefulness of case-based experience reusein production.Findings – This paper presents a concept of case-based experience reuse in production. Amaintenance agent using a Case-Based Reasoning approach to collect, preserve and reuseavailable experience in the form of sound recordings exemplifies this concept. Sound fromnormal and faulty robot gearboxes are recorded during the production end test and stored in acase library together with their diagnosis results. Given an unclassified sound signal, relevantcases are retrieved to aid a human in the decision making process. The maintenance agentdemonstrated good performance by making right judgments in 91% of all the tests, which isbetter than an inexperienced technician.Originality/value – The main focus of this paper is to show how to perform efficientexperience reuse in modern production industry to improve quality of products. Twoapproaches are used: a case-study describing an example of experience reuse in productionusing a fault diagnosis system recognizing and diagnosing audible faults on industrial robotsand an efficient approach on how to package such a system using the agent paradigm and agent architecture.

Place, publisher, year, edition, pages
Emerald, 2009
Keyword
Experience Reuse, Decision Support Systems, Condition Monitoring, Intelligent Agents, Case-Based Reasoning, Quality Improvement
National Category
Computer Science
Research subject
Computer Science
Identifiers
urn:nbn:se:mdh:diva-6534 (URN)10.1108/13552510910961129 (DOI)2-s2.0-67650668459 (Scopus ID)
Projects
IntMaint
Available from: 2009-07-06 Created: 2009-07-06 Last updated: 2013-12-03Bibliographically approved
3. Fault Diagnosis in Industry Using Sensor Readings and Case-Based Reasoning
Open this publication in new window or tab >>Fault Diagnosis in Industry Using Sensor Readings and Case-Based Reasoning
2004 (English)In: Journal of Intelligent & Fuzzy Systems, ISSN 1064-1246, Vol. 15, no 1, 41-46 p.Article in journal (Refereed) Published
Abstract [en]

Fault diagnosis of industrial equipments becomes increasingly important for improving the quality of manufacturing and reducing the cost for product testing. Developing a fast and reliable diagnosis system presents a challenge issue in many complex industrial scenarios. The major difficulties therein arise from contaminated sensor readings caused by heavy background noise as well as the unavailability of experienced technicians for support. In this paper we propose a novel method for diagnosis of faults by means of case-based reasoning and signal processing. The received sensor signals are processed by wavelet analysis to filter out noise and at the same time to extract a group of related features that constitutes a reduced representation of the original signal. The derived feature vector is then forwarded to a classification component that uses case-based reasoning to recommend a fault class for the probe case. This recommendation is based on previously classified cases in a case library. Case-based diagnosis has attractive properties in that it enables reuse of past experiences whereas imposes no demand on the size of the case base. The proposed approach has been applied to fault diagnosis of industrial robots at ABB Robotics and the results of experiments are very promising.

National Category
Software Engineering Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:mdh:diva-2232 (URN)000226874700006 ()2-s2.0-10044228485 (Scopus ID)
Available from: 2008-11-11 Created: 2008-11-11 Last updated: 2015-06-29Bibliographically approved
4. Fault Diagnosis of Industrial Robots using Acoustic Signals and Case-Based Reasoning
Open this publication in new window or tab >>Fault Diagnosis of Industrial Robots using Acoustic Signals and Case-Based Reasoning
2004 (English)In: Case-Based Reasoning. ECCBR 2004. Lecture Notes in Computer Science, vol 3155, 2004, 686-701 p.Conference paper, Published paper (Refereed)
Abstract [en]

In industrial manufacturing rigorous testing is used to ensure that the delivered products meet their specifications. Mechanical maladjustment or faults often show their presence as deviations compared to a normal sound pro-file. This is the case in robot assembly, the selected application domain for the system. Manual diagnosis based on sound requires extensive experience, and the experience is often acquired through costly mistakes and reduced production efficiency or quality loss caused by missed faults. The acquired experience is also difficult to preserve and transfer, and often lost if personnel leave the task of testing and fault diagnosis. We propose a Case-Based Reasoning approach to collect and preserve experience. The solution enables fast experience transfer and leads to less experienced personnel required to make more reliable and informed testing. Sounds from normal and faulty equipment are recorded and stored in a case library together with a diagnosis. Addition of new validated sound profiles continuously improves the system’s performance. The system can preserve and transfer experience between technicians, reducing overall fault identification time and increases quality by reduced number of missed faults. The original sound recordings are stored in form of the extracted features to-gether with other experience, e.g. instructions, additional tests, advice, user feedback etc.

Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 3155
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:mdh:diva-2241 (URN)10.1007/978-3-540-28631-8_50 (DOI)
Conference
European Conference on Case-Based Reasoning ECCBR 2004
Available from: 2008-11-11 Created: 2008-11-11 Last updated: 2017-02-10Bibliographically approved
5. Dynamic Modeling and Sound (Noise) Diagnostics of Robot Gearboxes for Fault Assessments
Open this publication in new window or tab >>Dynamic Modeling and Sound (Noise) Diagnostics of Robot Gearboxes for Fault Assessments
2005 (English)In: Proceedings of SIMS 2005 - Scandinavian Conference on Simulation and Modeling, 2005Conference paper, Published paper (Refereed)
Abstract [en]

Some gear faults in industrial robots can during operation be recognized as abnormal noise peaks coming from the gearbox. A library of such recordings has been assembled in order to automate fault diagnosis of the robots. A computer records sound from the gearbox and compare the new recordings with recordings stored in the library. The result of the comparison is a diagnosis of the condition of the robot. This paper proposes an extension of the sound library by incorporating model based reasoning. A dynamic model of the gearbox in the drive system has been constructed and gear vibrations on the force level are extracted from the model. These vibrations are projected onto the sound recordings with a statistical vibration diagnostic parameter known as the Crest Factor CF.

National Category
Other Engineering and Technologies
Identifiers
urn:nbn:se:mdh:diva-2736 (URN)
Conference
SIMS 2005 - Scandinavian Conference on Simulation and Modeling
Available from: 2008-11-11 Created: 2008-11-11 Last updated: 2015-10-12Bibliographically approved
6. Identifying Discriminating Features in Time Series Data for Diagnosis of Industrial Machines
Open this publication in new window or tab >>Identifying Discriminating Features in Time Series Data for Diagnosis of Industrial Machines
2007 (English)In: The 24th annual workshop of the Swedish Artificial Intelligence Society, May, 2007, Boras, Sweden,, 2007Conference paper, Published paper (Refereed)
Abstract [en]

Reducing the inherent high dimensionality in time series data is a desirable goal. Algorithms used for classi¯cation can easily be misguided if presented with data of to high dimension. E.g. the k-nearest neighbor algorithm which is often used for case-based classi¯cation per- forms best on smaller dimensions with less than 20 attributes. In this paper we address the problem using a time series case base and a feature discrimination approach incorporating an unsupervised combination of a search function based on statistical feature discrimination and a crite- rion function ¯nding the global maximum of discriminating power in the range the search function. Feature vectors for case indexing is computed with respect to this information. For evaluation, previously classi¯ed cur- rent measurements from an electrical motor driving the gearbox of axis 4 on an industrial robot were used. The results were promising and we managed to correctly classify measurements from healthy and unhealthy gearboxes.

Identifiers
urn:nbn:se:mdh:diva-2740 (URN)
Available from: 2008-11-11 Created: 2008-11-11 Last updated: 2014-06-26Bibliographically approved

Open Access in DiVA

fulltext(3116 kB)3214 downloads
File information
File name FULLTEXT01.pdfFile size 3116 kBChecksum SHA-512
f4d187904ea98d5705358fae65adde6c7cf823a80556075a01beefacd90f2aaf354d702e0423df28c470b79bf37417500cffdb3c903e9e887b852d601cb0b9fd
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Olsson, Erik
By organisation
School of Innovation, Design and Engineering
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 3214 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

Total: 464 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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