Real-time equipment condition assessment for a class-imbalanced dataset based on heterogeneous ensemble learning

• Autorzy: Chen Xiaohui , Zhang Zhiyao , Zhang Ze

Identyfikatory

DOI

10.17531/ein.2019.1.9

Warianty tytułu

PL

Ocena stanu sprzętu w czasie rzeczywistym dla zbiorów danych o niezrównoważonym rozkładzie w klasach. Metoda oparta na uczeniu zespołowym

• Języki publikacji: EN

Abstrakty

EN

This study proposes an ensemble learning model for the purpose of performing a real-time equipment condition assessment. This model makes it possible to plan desired preventive maintenance activities before an unexpected failure takes place. This study focuses on the class-imbalanced problem in equipment condition assessment research. In reality, equipment will experience multiple conditions(states), most of the time remaining in the normal condition and relatively rarely being in the critical condition, which means that, from the perspective of data modelling, the distribution of samples is highly imbalanced among different classes(conditions). The majority of samples belong to the normal condition, while the minority belong to the critical condition, which poses a great challenge to the classification performance. To address this problem, a genetic algorithm-based ensemble learning model is presented. Furthermore, a self-updating learning strategy is presented for online monitoring, contributing to adaptability and reliability enhancement along with time. Many previous studies have attempted feature extraction and to set thresholds for equipment health indicators. This study has an advantage of omitting these steps, as it can directly assess the equipment condition through the proposed ensemble learning model. Numerical experiments, including two types of comparison studies, have been conducted. The results show the greater effectiveness of our proposed model over that of previous research in terms of the stability and accuracy of its classification performance.

PL

W pracy przedstawiono model uczenia maszynowego opartego na zespołach niejednorodnych klasyfikatorów (ensemble learning), który pozwala przeprowadzać ocenę stanu sprzętu w czasie rzeczywistym. Model ten umożliwia zaplanowanie niezbędnych czynności konserwacji profilaktycznej przed wystąpieniem niespodziewanego uszkodzenia. Tematem pracy jest zagadnienie niezrównoważonego rozkładu w klasach poruszane w badaniach dotyczących oceny stanu sprzętu. W warunkach rzeczywistych, sprzęt chrakteryzuje wiele różnych stanów, przy czym przez większość czasu pozostaje on w stanie normalnym, a relatywnie rzadko znajduje się w stanie krytycznym, co oznacza, że z punktu widzenia modelowania danych, rozkład prób w poszczególnych klasach (stanach) jest wysoce niezrównoważony. Większość prób należy do stanu normalnego, a mniejszość do stanu krytycznego, co stanowi duże wyzwanie jeśli chodzi o wydajność klasyfikacji. W celu rozwiązania tego problemu, przedstawiono model uczenia zespołowego oparty na algorytmie genetycznym. Ponadto zaprezentowano samoaktualizującą się strategię uczenia wykorzystywaną do monitorowania online, która wraz z upływem czasu zwiększa adaptacyjność i niezawodność modelu . W wielu poprzednich badaniach podejmowano próby ekstrakcji cech oraz ustalania progów dla wskaźników stanu sprzętu. Zaletą przedstawionej metody jest to, że pozwala ona pominąć te etapy i bezpośrednio oceniać stan sprzętu za pomocą proponowanego modelu uczenia zespołowego. Przeprowadzono eksperymenty numeryczne, w tym dwa rodzaje badań porównawczych. Wyniki pokazują większą skuteczność proponowanego modelu w stosunku do poprzednich badań pod względem stabilności i trafności klasyfikacji.

Słowa kluczowe

EN

condition assessment; heterogeneous ensemble learning; genetic algorithm; class-imbalanced

PL

ocena stanu; uczenie zespołowe; algorytm genetyczny; niezrównoważony rozkład w klasach

• Wydawca: Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne
• Czasopismo: Eksploatacja i Niezawodność
• Rocznik: 2019
• Tom: Vol. 21, no. 1
• Strony: 68--80
• Opis fizyczny: Bibliogr. 41 poz., rys., tab.

Twórcy

autor

Chen Xiaohui

• The State Key Lab of Mechanical Transmission Chongqing University Chongqing, China

autor

Zhang Zhiyao

• The State Key Lab of Mechanical Transmission Chongqing University Chongqing, China

autor

Zhang Ze

• The State Key Lab of Mechanical Transmission Chongqing University Chongqing, China

Bibliografia

• 1. Albisua I, Arbelaitz O, Gurrutxaga I, Lasarguren A, Muguerza J, Pérez JM. The quest for the optimal class distribution: an approach for enhancing the effectiveness of learning via resampling methods for imbalanced data sets. Progress in Artificial Intelligence 2013; 2(1): 45- 63, https://doi.org/10.1007/s13748-012-0034-6.
• 2. Amir M D M, Muttalib E S A, editors. Health index assessment of aged oil-filled ring main units. Power Engineering and Optimization Conference; 2014, https://doi.org/10.1109/PEOCO.2014.6814452.
• 3. Baik H-S, Jeong H S, Abraham D M. Estimating transition probabilities in Markov chain-based deterioration models for management of wastewater systems. Journal of water resources planning and management 2006; 132(1): 15-24, https://doi.org/10.1061/(ASCE)0733- 9496(2006)132:1(15).
• 4. Benkedjouh T, Medjaher K, Zerhouni N, Rechak S. Health assessment and life prediction of cutting tools based on support vector regression. Journal of Intelligent Manufacturing 2015; 26(2): 213-223, https://doi.org/10.1007/s10845-013-0774-6.
• 5. Bennin K E, Keung J, Phannachitta P, Monden A, Mensah S. Mahakil: Diversity based oversampling approach to alleviate the class imbalance issue in software defect prediction. IEEE Transactions on Software Engineering 2018; 44(6), https://doi.org/10.1109/TSE.2017.2731766.
• 6. Caruana R, Niculescu-Mizil A, Crew G, Ksikes A, editors. Ensemble selection from libraries of models. International Conference on Machine Learning 2004, https://doi.org/10.1145/1015330.1015432.
• 7. Carvalho E, Tang F, Allen E, Sharma P, editors. A Case Study of Asset Integrity and Risk Assessment for Subsea Facilities and Equipment Life Extension. Offshore Technology Conference 2015, https://doi.org/10.4043/25701-MS.
• 8. Chan C W, Paelinckx D. Evaluation of Random Forest and Adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery. Remote Sensing of Environment 2008; 112(6): 2999-3011, https://doi.org/10.1016/j. rse.2008.02.011.
• 9. Chawla N V, Bowyer K W, Hall LO, Kegelmeyer W P. SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research 2002; 16: 321-357, https://doi.org/10.1613/jair.953.
• 10. Cheng F, Zhang J, Wen C. Cost-Sensitive Large margin Distribution Machine for classification of imbalanced data. Pattern Recognition Letters 2016; 80: 107-112, https://doi.org/10.1016/j.patrec.2016.06.009.
• 11. Fan M, Zeng Z, Zio E, Kang R. Modeling dependent competing failure processes with degradation-shock dependence. Reliability Engineering & System Safety 2017; 165: 422-430, https://doi.org/10.1016/j.ress.2017.05.004.
• 12. Fan M, Zeng Z, Zio E, Kang R, Chen Y. A stochastic hybrid systems based framework for modeling dependent failure processes. PloS one 2017; 12(2), https://doi.org/10.1371/journal.pone.0172680.
• 13. Giorgio M, Guida M, Pulcini G. An age- and state-dependent Markov model for degradation processes. IIE Transactions 2011; 43(9): 621- 632, https://doi.org/10.1080/0740817X.2010.532855.
• 14. Haixiang G, Yijing L, Shang J, Mingyun G, Yuanyue H, Bing G. Learning from class-imbalanced data: Review of methods and applications. Expert Systems with Applications 2017; 73: 220-239, https://doi.org/10.1016/j.eswa.2016.12.035.
• 15. Haque M N, Noman N, Berretta R, Moscato P. Heterogeneous Ensemble Combination Search Using Genetic Algorithm for Class Imbalanced Data Classification. Plos One 2016; 11(1): e0146116, https://doi.org/10.1371/journal.pone.0146116.
• 16. Hastie T, Tibshirani R, Friedman J. Ensemble Learning: Springer New York; 2009. 605-624.
• 17. Japkowicz N. Learning from Imbalanced Data Sets: A Comparison of Various Strategies. Aaai Workshop on Learning from Imbalanced Data Sets 2000: 10-15.
• 18. Kleiner Y, Sadiq R, Rajani B B. Modeling Failure Risk in Buried Pipes Using Fuzzy Markov Deterioration Process. Pipeline Engineering and Construction@sWhat's on the Horizon; 2004.
• 19. Lee W, Jun C-H, Lee J-S. Instance categorization by support vector machines to adjust weights in AdaBoost for imbalanced data classification. Information Sciences 2017; 381: 92-103, https://doi.org/10.1016/j.ins.2016.11.014.
• 20. Li Z, Zhang B, Wang Y, Chen F, Taib R, Whiffin V, et al. Water pipe condition assessment: a hierarchical beta process approach for sparse incident data. Machine Learning 2014; 95(1): 11-26, https://doi.org/10.1007/s10994-013-5386-z.
• 21. López A J G, Márquez A C, Macchi M, Fernández JFG. Prognostics and Health Management in Advanced Maintenance Systems. Advanced Maintenance Modelling for Asset Management: Springer; 2018. 79-106, https://doi.org/10.1007/978-3-319-58045-6_4.
• 22. López V, Fernández A, García S, Palade V, Herrera F. An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics. Information Sciences 2013; 250: 113-141, https://doi.org/10.1016/j.ins.2013.07.007.
• 23. López V, Fernández A, Herrera F. On the importance of the validation technique for classification with imbalanced datasets: Addressing covariate shift when data is skewed. Information Sciences 2014; 257: 1-13, https://doi.org/10.1016/j.ins.2013.09.038.
• 24. Luengo J, Fernández A, García S, Herrera F. Addressing data complexity for imbalanced data sets: analysis of SMOTE-based oversampling and evolutionary undersampling. Soft Computing 2011; 15(10): 1909-1936, https://doi.org/10.1007/s00500-010-0625-8.
• 25. Huk M, Szczepanik M, Multiple classifier error probability for multi-clas problems. Eksploatacja i Niezawodnosc - Maintenance and Reliability 2011; 51(3): 12-16.
• 26. Margineantu D D, Dietterich T G, editors. Pruning Adaptive Boosting. Fourteenth International Conference on Machine Learning; 1997
• 27. Mathew J, Pang C K, Luo M, Leong W H. Classification of imbalanced data by oversampling in kernel space of support vector machines. IEEE Transactions on Neural Networks and Learning Systems 2018; 29(9): 4065-4076, https://doi.org/10.1109/TNNLS.2017.2751612.
• 28. Oreški Stjepan, Oreški Goran. Cost-Sensitive Learning from Imbalanced Datasets for Retail Credit Risk Assessment. TEM JOURNAL - Technology, Education, Management, Informatics 2018.
• 29. Othman A, Tahir M, El Shatshat R, Shaban K, editors. Application of ensemble classification method for power transformers condition assessment. Electrical and Computer Engineering (CCECE), 2017 IEEE 30th Canadian Conference; 2017: IEEE.
• 30. Parradohernández E, Robles G, Ardilarey J A, Martíneztarifa J M, Sciubba E. Robust Condition Assessment of Electrical Equipment with One Class Support Vector Machines Based on the Measurement of Partial Discharges. Energies 2018; 11(3).
• 31. Partalas I, Tsoumakas G, Vlahavas I, editors. Focused Ensemble Selection: A Diversity-Based Method for Greedy Ensemble Selection. Conference on ECAI 2008: European Conference on Artificial Intelligence; 2008.
• 32. Pedregosa F, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, et al. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research 2011; 12(10): 2825-2830.
• 33. Razi-Kazemi A A, Vakilian M, Niayesh K, Lehtonen M. Data Mining of Online Diagnosed Waveforms for Probabilistic Condition Assessment of SF6 Circuit Breakers. IEEE Transactions on Power Delivery 2015; 30(3): 1354-1362, https://doi.org/10.1109/TPWRD.2015.2399454.
• 34. Reddy M V, Sodhi R. A rule-based S-Transform and AdaBoost based approach for power quality assessment. Electric Power Systems Research 2016; 134: 66-79, https://doi.org/10.1016/j.epsr.2016.01.003.
• 35. Sugier J, Anders G J. Modelling and evaluation of deterioration process with maintenance activities. Eksploatacja i Niezawodnosc - Maintenance and Reliability 2013; 15(4): 305-311.
• 36. Tseng M-L, Wu K-J, Ma L, Kuo TC, Sai F. A hierarchical framework for assessing corporate sustainability performance using a hybrid fuzzy synthetic method-DEMATEL. Technological Forecasting and Social Change 2017, https://doi.org/10.1016/j.techfore.2017.10.014.
• 37. Xu J, Xu L. Integrated system health management-based condition assessment for manned spacecraft avionics. Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering 2013; 227(1): 19-32, https://doi.org/10.1177/0954410011431395.
• 38. Zadrozny B, Langford J, Abe N, editors. Cost-sensitive learning by cost-proportionate example weighting. Data Mining, 2003 ICDM 2003 Third IEEE International Conference on; 2003.
• 39. Zhang L, Zhang J, Zhai H, Zhou S. A new assessment method of mechanism reliability based on chance measure under fuzzy and random uncertainties. Eksploatacja i Niezawodnosc - Maintenance and Reliability 2018; 20(2): 219-228, https://doi.org/10.17531/ein.2018.2.06.
• 40. Zhu L, Lu C, Dong Z Y, Hong C. Imbalance Learning Machine-Based Power System Short-Term Voltage Stability Assessment. IEEE Transactions on Industrial Informatics 2017; 13 (5): 2533-2543, https://doi.org/10.1109/TII.2017.2696534
• 41. Zhu L J, Cong H. The State Assessment of Armored Vehicle Engine Based on Analytic Hierarchy Process and Fuzzy Synthetic Evaluation. Advanced Materials Research 2014; 988(988): 606-610, https://doi.org/10.4028/www.scientific.net/AMR.988.606.

Uwagi

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).