Change search
Link to record
Permanent link

Direct link
BETA
Alternative names
Publications (10 of 63) Show all publications
Zhang, L., Lin, J. & Karim, R. (2018). Adaptive Kernel Density-based Anomaly Detection for Nonlinear Systems. Knowledge-Based Systems, 139(1), 50-63
Open this publication in new window or tab >>Adaptive Kernel Density-based Anomaly Detection for Nonlinear Systems
2018 (English)In: Knowledge-Based Systems, ISSN 0950-7051, E-ISSN 1872-7409, Vol. 139, no 1, p. 50-63Article in journal (Refereed) Published
Abstract [en]

This paper presents an unsupervised, density-based approach to anomaly detection. The purpose is to define a smooth yet effective measure of outlierness that can be used to detect anomalies in nonlinear systems. The approach assigns each sample a local outlier score indicating how much one sample deviates from others in its locality. Specifically, the local outlier score is defined as a relative measure of local density between a sample and a set of its neighboring samples. To achieve smoothness in the measure, we adopt the Gaussian kernel function. Further, to enhance its discriminating power, we use adaptive kernel width: in high-density regions, we apply wide kernel widths to smooth out the discrepancy between normal samples; in low-density regions, we use narrow kernel widths to intensify the abnormality of potentially anomalous samples. The approach is extended to an online mode with the purpose of detecting anomalies in stationary data streams. To validate the proposed approach, we compare it with several alternatives using synthetic datasets; the approach is found superior in terms of smoothness, effectiveness and robustness. A further experiment on a real-world dataset demonstrated the applicability of the proposed approach in fault detection tasks.

Place, publisher, year, edition, pages
Elsevier, 2018
Keywords
maintenance modelling, fault detection, unsupervised learning, nonlinear data, kernel density
National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-60425 (URN)10.1016/j.knosys.2017.10.009 (DOI)000417773400005 ()2-s2.0-85031504017 (Scopus ID)
Note

Validerad;2017;Nivå 2;2017-11-21 (andbra)

Available from: 2016-11-15 Created: 2016-11-15 Last updated: 2017-12-28Bibliographically approved
Cai, B., Huang, L., Lin, J. & Xie, M. (2017). Bayesian Networks in Fault Diagnosis: some research issues andchallenges. In: Diego Galar, Dammika Seneviratne (Ed.), Proceedings of MPMM 2016: 6th International Conference on Maintenance Performance Measurement and Management, 28 November 2016, Luleå, Sweden. Paper presented at Maintenance Performance and Measurement and Management 2016(MPMM 2016). November 28, Luleå, Sweden (pp. 26-32). Luleå: Luleå tekniska universitet
Open this publication in new window or tab >>Bayesian Networks in Fault Diagnosis: some research issues andchallenges
2017 (English)In: Proceedings of MPMM 2016: 6th International Conference on Maintenance Performance Measurement and Management, 28 November 2016, Luleå, Sweden / [ed] Diego Galar, Dammika Seneviratne, Luleå: Luleå tekniska universitet, 2017, p. 26-32Conference paper, Published paper (Refereed)
Place, publisher, year, edition, pages
Luleå: Luleå tekniska universitet, 2017
National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-63486 (URN)978-91-7583-841-0 (ISBN)
Conference
Maintenance Performance and Measurement and Management 2016(MPMM 2016). November 28, Luleå, Sweden
Available from: 2017-05-22 Created: 2017-05-22 Last updated: 2017-11-24Bibliographically approved
Yu, H., Yang, J., Lin, J. (. & Zhao, Y. (2017). Reliability evaluation of non-repairable phased-mission common bus systems with common cause failures. Computers & industrial engineering, 111, 445-457
Open this publication in new window or tab >>Reliability evaluation of non-repairable phased-mission common bus systems with common cause failures
2017 (English)In: Computers & industrial engineering, ISSN 0360-8352, E-ISSN 1879-0550, Vol. 111, p. 445-457Article in journal (Refereed) Published
Abstract [en]

Phased-mission common bus (PMCB) systems are systems with a common bus structure, performing missions with consecutive and non-overlapping phases of operations. PMCB systems are found throughout industry, e.g., power generating systems, parallel computing systems, transportation systems, and are sometimes characterized by their common cause failures. Reliability evaluation of PMCB systems plays an important role in system design, operation, and maintenance. However, current studies have focused on either phased-mission systems or common bus systems because of their complexity. The challenge in practice is to consider phased-mission systems together with common bus structures and common cause failures. To solve this problem, we propose an evaluation algorithm for PMCB systems with common cause failures by coupling the structure function of a common bus performance sharing system and an existing recursive algorithm. To weigh the efficiency of the proposed algorithm, its complexity is discussed. To improve the reliability of PMCB systems, we adopt the genetic algorithm method to search for the optimal allocation strategies of the service elements. We use both analytical and numerical examples to illustrate the application of the proposed algorithm.

Place, publisher, year, edition, pages
Elsevier, 2017
National Category
Engineering and Technology Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-65001 (URN)10.1016/j.cie.2017.08.002 (DOI)000410468600036 ()2-s2.0-85027888677 (Scopus ID)
Note

Validerad;2017;Nivå 2;2017-08-16 (rokbeg)

Available from: 2017-08-10 Created: 2017-08-10 Last updated: 2018-07-10Bibliographically approved
Zhang, L., Lin, J. & Karim, R. (2017). Sliding Window-based Fault Detection from High-dimensional Data Streams (ed.). IEEE Transactions on Systems, Man & Cybernetics. Systems, 47(2), 289-303, Article ID 7509594.
Open this publication in new window or tab >>Sliding Window-based Fault Detection from High-dimensional Data Streams
2017 (English)In: IEEE Transactions on Systems, Man & Cybernetics. Systems, ISSN 2168-2216, Vol. 47, no 2, p. 289-303, article id 7509594Article in journal (Refereed) Published
Abstract [en]

High-dimensional data streams are becoming increasingly ubiquitous in industrial systems. Efficient detection of system faults from these data can ensure the reliability and safety of the system. The difficulties brought about by high dimensionality and data streams are mainly the ``curse of dimensionality'' and concept drifting, and one current challenge is to simultaneously address them. To this purpose, this paper presents an approach to fault detection from nonstationary high-dimensional data streams. An angle-based subspace anomaly detection approach is proposed to detect low-dimensional subspace faults from high-dimensional datasets. Specifically, it selects fault-relevant subspaces by evaluating vectorial angles and computes the local outlier-ness of an object in its subspace projection. Based on the sliding window strategy, the approach is further extended to an online mode that can continuously monitor system states. To validate the proposed algorithm, we compared it with the local outlier factor-based approaches on artificial datasets and found the algorithm displayed superior accuracy. The results of the experiment demonstrated the efficacy of the proposed algorithm. They also indicated that the algorithm has the ability to discriminate low-dimensional subspace faults from normal samples in high-dimensional spaces and can be adaptive to the time-varying behavior of the monitored system. The online subspace learning algorithm for fault detection would be the main contribution of this paper.

National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-12680 (URN)10.1109/TSMC.2016.2585566 (DOI)000396231100008 ()2-s2.0-85009962444 (Scopus ID)bd869fb2-6668-49d9-9511-535c94003f4f (Local ID)bd869fb2-6668-49d9-9511-535c94003f4f (Archive number)bd869fb2-6668-49d9-9511-535c94003f4f (OAI)
Note

Validerad; 2017; Nivå 2; 2017-01-23 (andbra)

Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2018-07-10Bibliographically approved
Asplund, M., Lin, J. & Rantatalo, M. (2016). Assessment of the data quality of wayside wheel profile measurements (ed.). International Journal of COMADEM, 19(3), 19-25
Open this publication in new window or tab >>Assessment of the data quality of wayside wheel profile measurements
2016 (English)In: International Journal of COMADEM, ISSN 1363-7681, Vol. 19, no 3, p. 19-25Article in journal (Refereed) Published
Abstract [en]

To evaluate the behaviour and the condition of a railway wheel in relation to performance and safety criteria, the wheel profile can be measured. This can be achieved using manual methods or automatic systems mounted along the railway track. Such systems have the advantage that they can measure a vast number of profiles, enabling new possibilities of performing statistical analyses of the results and pinpointing bad wheels at an early stage. These wayside measurement systems are, however, subjected to different environmental conditions that can affect the data quality of the measurement. If one is to be able to use automatic wheel profile measurements, the data quality has to be controlled in order to facilitate maintenance decisions. This paper proposes a method for the data quality assessment of an automatic wayside condition monitoring system measuring railway rolling stock wheels. The purpose of the assessment method proposed in this paper is to validate individual wheel profile measurements to ensure the accuracy of the wheel profile measurement data and hence the following data analysis. The method consists of a check routine based on the paired t-test, which uses a hypothesis test to verify if the null hypotheses are true. The check routine compares measurements of passing wheels rolling to a certain destination with measurements of the same wheels returning from that destination. The routine of comparing measurements of the same wheel, which is performed by four sensors (one on each side of each rail), will ensure that the sensors generate the same data for the same sample. A case study is presented which shows how the method can detect a faulty setup of the measurement system and prevent incorrect interpretations of the data from different measurement units in the same system. The paper ends with a discussion and conclusions concerning the improvements that are presented.

National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-10887 (URN)2-s2.0-84987680545 (Scopus ID)9c3cbb22-ce80-4a35-8f12-e4f456f6a51b (Local ID)9c3cbb22-ce80-4a35-8f12-e4f456f6a51b (Archive number)9c3cbb22-ce80-4a35-8f12-e4f456f6a51b (OAI)
Note

Validerad; 2016; Nivå 1; 2016-10-26 (inah)

Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2017-11-24Bibliographically approved
Lin, J. (2016). Bayesian Reliability with MCMC: Opportunities and Challenges (ed.). In: (Ed.), Uday Kumar; Alireza Ahmadi; Ajit Kumar Verma; Prabhakar Varde (Ed.), Current Trends in Reliability, Availability, Maintainability and Safety: An Industry Perspective. Paper presented at International Conference ICRESH-ARMS 2015 : 01/06/2015 - 04/06/2015 (pp. 575-585). Encyclopedia of Global Archaeology/Springer Verlag
Open this publication in new window or tab >>Bayesian Reliability with MCMC: Opportunities and Challenges
2016 (English)In: Current Trends in Reliability, Availability, Maintainability and Safety: An Industry Perspective / [ed] Uday Kumar; Alireza Ahmadi; Ajit Kumar Verma; Prabhakar Varde, Encyclopedia of Global Archaeology/Springer Verlag, 2016, p. 575-585Conference paper, Published paper (Refereed)
Abstract [en]

The recent proliferation of Markov Chain Monte Carlo (MCMC) approaches has led to the use of the Bayesian inference in a wide variety of fields, including reliability engineering. With the current (and future) proliferation of new products, old problems continue to hamper us, while new challenges keep appearing. In Bayesian reliability, these include but are not limited to: (1) achieving and making use of prior information; (2) applying small data sets or system operating/environmental (SOE) data with big and complex data; and (3) making posterior inferences from high-dimensional numerical integration. To deal with old problems while meeting new challenges, this paper proposes an improved procedure for Bayesian reliability inference with MCMC, discussing modern reliability data and noting some applications where the Bayesian reliability approach with MCMC can be used. It also explores opportunities to use Bayesian reliability models to create stronger statistical methods from prior to posterior. Finally, it outlines some practical concerns and remaining challenges for future research.

Place, publisher, year, edition, pages
Encyclopedia of Global Archaeology/Springer Verlag, 2016
Series
Lecture Notes in Mechanical Engineering, ISSN 2195-4356
National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-39619 (URN)10.1007/978-3-319-23597-4_41 (DOI)2-s2.0-85043794030 (Scopus ID)e7259d28-6d42-4c73-b563-f373b27ea506 (Local ID)978-3-319-23596-7 (ISBN)978-3-319-23597-4 (ISBN)e7259d28-6d42-4c73-b563-f373b27ea506 (Archive number)e7259d28-6d42-4c73-b563-f373b27ea506 (OAI)
Conference
International Conference ICRESH-ARMS 2015 : 01/06/2015 - 04/06/2015
Note

Godkänd; 2016; Bibliografisk uppgift: Containing selected papers from the ICRESH-ARMS 2015 conference in Lulea, Sweden, collected by editors with years of experiences in Reliability and maintenance modeling, risk assessment, and asset management, this work maximizes reader insights into the current trends in Reliability, Availability, Maintainability and Safety (RAMS) and Risk Management. Featuring a comprehensive analysis of the significance of the role of RAMS and Risk Management in the decision making process during the various phases of design, operation, maintenance, asset management and productivity in Industrial domains, these proceedings discuss key issues and challenges in the operation, maintenance and risk management of complex engineering systems and will serve as a valuable resource for those in the field. ; 20151223 (andbra)

Available from: 2016-10-03 Created: 2016-10-03 Last updated: 2018-03-29Bibliographically approved
Lin, J., Nordmark, T. & Zhang, L. (2016). Data analysis of heavy haul wagon axle loads on Malmbanan line, Sweden: A case study for LKAB (ed.). Paper presented at . Luleå: Luleå tekniska universitet
Open this publication in new window or tab >>Data analysis of heavy haul wagon axle loads on Malmbanan line, Sweden: A case study for LKAB
2016 (English)Report (Other academic)
Abstract [en]

The research presented in this report was carried out by Operation and Maintenance Engineering at Luleå University of Technology (LTU) from November 2015 to April 2016. LKAB initiated the research study and provided financial support. The purpose of this research was to support LKAB and Trafikverket in their operational strategy review and optimization of future axle load implementations. It developed five research questions and answered them by analyzing the data for the Malmbanan iron ore train axle loads for 2015.Data analysis comprises four parts. In the first part (section 2), the analysis focuses on axle loads of all loaded trains operating at three different terminals: Kiruna, Malmberget, and Svappavaara. In addition, it examines the differences of three weighing locations in Kiruna, five weighing locations in Malmberget and four weighing locations in Svappavaara (12 weighing locations). Based on these results, the analysis in the second part (section 3) focuses on the heavy haul wagon. Wagon loads are evaluated and predicted for different loading rules (31.0 and 32.5 tons separately). To optimize the current loading rules, the third part of the analysis (section 4) proposes a novel approach to optimize the wagon axle loads: “three sigma prediction”. Under this approach, Kiruna, Malmberget and Svappavaara can set new target loads based on various risk levels. In the fourth and final part of the data analysis (section 5), a comparison study is carried out by collecting axle load data for the test train (with a 32.5 ton axle load) using three different measurement systems in Malmberget, Sävast and Sunderbyn. Finally, sections 6 and 7 summarize the results and make some recommendations for future work. The work presented in this report should give LKAB and Trafikverket a good overview of the load distribution for the ore trains operating on Malmbanan line. It can serve as input into the process of evaluating possible changes in axle load limits. It also gives LKAB a base from which to identify and work with optimization of the various loading places to load trains more efficiently and save costs.

Place, publisher, year, edition, pages
Luleå: Luleå tekniska universitet, 2016
Series
Technical report / Luleå University of Technology, ISSN 1402-1536
National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-24113 (URN)9bbc981e-b5b8-440c-8399-56486fadd26f (Local ID)978-91-7583-593-8 (ISBN)978-91-7583-594-5 (ISBN)9bbc981e-b5b8-440c-8399-56486fadd26f (Archive number)9bbc981e-b5b8-440c-8399-56486fadd26f (OAI)
Note

Godkänd; 2016; 20160502 (linjan)

Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2017-11-24Bibliographically approved
Norrbin, P., Lin, J. & Parida, A. (2016). Energy efficiency optimization for railway switches & crossings: a case study in Sweden (ed.). Paper presented at World Congress of Railway Research : 29/05/2016 - 02/06/2016. Paper presented at World Congress of Railway Research : 29/05/2016 - 02/06/2016.
Open this publication in new window or tab >>Energy efficiency optimization for railway switches & crossings: a case study in Sweden
2016 (English)Conference paper, Oral presentation only (Refereed)
National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-37249 (URN)b3987a25-04e1-4ebd-bb24-869f79896376 (Local ID)b3987a25-04e1-4ebd-bb24-869f79896376 (Archive number)b3987a25-04e1-4ebd-bb24-869f79896376 (OAI)
Conference
World Congress of Railway Research : 29/05/2016 - 02/06/2016
Note
Godkänd; 2016; 20160523 (pernor)Available from: 2016-10-03 Created: 2016-10-03 Last updated: 2017-11-25Bibliographically approved
Asplund, M. & Lin, J. (2016). Evaluating the measurement capability of a wheel profile measurement system by using GR&R (ed.). Measurement, 92, 19-27
Open this publication in new window or tab >>Evaluating the measurement capability of a wheel profile measurement system by using GR&R
2016 (English)In: Measurement, ISSN 0263-2241, E-ISSN 1873-412X, Vol. 92, p. 19-27Article in journal (Refereed) Published
Abstract [en]

Reliable data with less variation play a key role for acceptance of the usefulness of the measurement output of a wheel profile measurement system (WPMS) in a railway network. However, in practice, most studies are carried out without checking the reliability of data from such a system, which may lead to inappropriate maintenance strategies. To ensure the measurement capability of WPMS and to support robust maintenance in railway systems, this study has evaluated measurement data for the flange height, flange thickness, flange slope, and tread hollowing of rolling stock wheels by using gauge repeatability and reproducibility (GR&R). In this study, acceptance and rejection criteria for the precision-to-tolerance ratio (PTR), signal-to-noise ratio (SNR), and discrimination ratio (DR) have been employed to evaluate the measurement capabilities. For the purpose of illustration, we have implemented a new proposed approach. This approach involves both an analysis using graphs with four regions with a confidence interval (CI) of 95% and an analysis using a graph with three regions with only the predicted values; the latter type of graph represents an innovation made in this study. The results show that the measurements of the tread hollowing and flange slope are on an acceptable level, while those for the flange height and flange thickness have to be rejected as unacceptable. The action proposed to increase the quality of data on the flange height and flange thickness is to enhance the calibration of the WPMS. In conclusion, GR&R is a useful tool to evaluate the measurement capability of WPMS and to provide helpful support for maintenance decision making.

Abstract [en]

Reliable data with less variation play a key role for acceptance of the usefulness of the measurement output of a wheel profile measurement system (WPMS) in a railway network. However, in practice, most studies are carried out without checking the reliability of data from such a system, which may lead to inappropriate maintenance strategies. To ensure the measurement capability of WPMS and to support robust maintenance in railway systems, this study has evaluated measurement data for the flange height, flange thickness, flange slope, and tread hollowing of rolling stock wheels by using gauge repeatability and reproducibility (GR&R). In this study, acceptance and rejection criteria for the precision-to-tolerance ratio (PTR), signal-to-noise ratio (SNR), and discrimination ratio (DR) have been employed to evaluate the measurement capabilities. For the purpose of illustration, we have implemented a new proposed approach. This approach involves both an analysis using graphs with four regions with a confidence interval (CI) of 95% and an analysis using a graph with three regions with only the predicted values; the latter type of graph represents an innovation made in this study. This graph has the advantages that it can visualize three different levels of data quality in same figure, namely “unacceptable”, “acceptable” and “good”, and also include a number of measures without becoming unclear, which are features that have been missing in previous presentations. The results show that the measurements of the flange slope are on an acceptable level, while those for the flange height, flange thickness and tread hollowing have to be rejected as unacceptable. The action proposed for increasing the quality of data on the flange height, flange thickness and tread hollowing is to enhance the calibration of the WPMS. In conclusion, GR&R is a useful tool to evaluate the measurement capability of WPMS and to provide helpful support for maintenance decision making. This investigation also shows that there is good reason to be careful when selecting measures and when interpreting the results, since, for a certain wheel profile parameter, when one measure is used, the results may be acceptable, but when another measure is used, the results may be unacceptable.

National Category
Other Civil Engineering
Research subject
Operation and Maintenance
Identifiers
urn:nbn:se:ltu:diva-9409 (URN)10.1016/j.measurement.2016.05.090 (DOI)000380736700003 ()2-s2.0-84974560549 (Scopus ID)8062b662-b2f7-4a53-9365-fe7966466b08 (Local ID)8062b662-b2f7-4a53-9365-fe7966466b08 (Archive number)8062b662-b2f7-4a53-9365-fe7966466b08 (OAI)
Note

Validerad; 2016; Nivå 2; 20160403 (matasp)

Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2018-07-10Bibliographically approved
Norrbin, P., Lin, J. & Parida, A. (2016). Infrastructure robustness for railway systems (ed.). International Journal of Pedagogy, Innovation and New Technologies, 12(3), 249-264, Article ID 5.
Open this publication in new window or tab >>Infrastructure robustness for railway systems
2016 (English)In: International Journal of Pedagogy, Innovation and New Technologies, ISSN 0973-1318, E-ISSN 2392-0092, Vol. 12, no 3, p. 249-264, article id 5Article in journal (Refereed) Published
Abstract [en]

In the railway industry, most maintenance approaches are based on certain “specified conditions”, e.g., RAMS (Reliability, Availability, Maintainability and Safety) and Risk. But the reality is more complex. Instead of the assumed conditions, “unfavorable conditions” may occur from either natural or operational causes, where robustness can be an effective approach. To adequately consider “unfavorable conditions” and to reduce “uncertainties” in railway maintenance, this study conducts a holistic examination of railway infrastructure robustness. It gives an overview of robustness and discusses some relevant studies. It then develops a new road map for railway infrastructure robustness, including a novel definition and a new framework of robustness management, based on continuous improvement. It explores the opportunities of applying the road map to the infrastructure of railway systems and outlines some practical concerns and remaining challenges for future research. The results provide guidelines for other research into robust infrastructure in railway maintenance.

National Category
Other Civil Engineering
Research subject
Operation and Maintenance; Sustainable transportation (AERI)
Identifiers
urn:nbn:se:ltu:diva-9983 (URN)2-s2.0-84991508605 (Scopus ID)8b91769d-0e85-47cb-9bc1-f8984a825597 (Local ID)8b91769d-0e85-47cb-9bc1-f8984a825597 (Archive number)8b91769d-0e85-47cb-9bc1-f8984a825597 (OAI)
Note

Validerad; 2016; Nivå 1; 20160523 (pernor)

Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2017-11-24Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-7458-6820

Search in DiVA

Show all publications