satНАУКА и ТЕХНИКАScience & Technique2227-10312414-0392Belarusian National Technical University10.21122/2227-1031-2019-18-6-519-524sat-2231Research ArticleМатериалы 16-го Европейского автомобильного конгрессаProceedings of the 16th European Automotive CongressПредварительный выбор сигнала для мониторинга и прогнозирования старения компонентов силовой передачи автомобиляSignal Pre-Selection for Monitoring and Prediction of Vehicle Powertrain Component AgingУдо СассА.Udo SassA.

Адрес для переписки: Удо Сасс Андреас – АО «Фольксваген» Бриффах 17772, 38436, г. Вольфсбург, Федеративная Республика Германия. Тел.: +7 916 940-00-06     andreas.udo.sass@volkswagen.de 

Address for correspondence: Udo Sass Andreas – Volkswagen AG, Brieffach 17772, 38436, Wolfsburg, Federal Republic of Germany.   Теl.: +7 916 940-00-06      andreas.udo.sass@volkswagen.de

andreas.udo.sass@volkswagen.deЭсатбейоглуЭ.EsatbeyogluE.ИвверксТ.IwwerksT.АО «Фольксваген»ГерманияVolkswagen AGGermany201905122019186519524Copyright © Удо Сасс А., Эсатбейоглу Э., Ивверкс Т., 20192019Удо Сасс А., Эсатбейоглу Э., Ивверкс Т.Udo Sass A., Esatbeyoglu E., Iwwerks T.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://sat.bntu.by/jour/article/view/2231

Прогнозное техническое обслуживание является важным для предотвращения незапланированных простоев современных транспортных средств. С расширением функциональности одновременно происходит быстрый рост обмена данными между электронными блоками управления. Большое количество бортовых сигналов позволяет осуществлять мониторинг процесса старения. Старение компонентов автомобиля зависит от того, как они используются. Элементы старения выявляются благодаря наличию ряда бортовых сигналов. В данной статье предложен метод выбора бортовых сигналов с целью определения соответствующих для проведения мониторинга и прогнозирования старения компонентов силовой передачи транспортных средств. Процесс старения рассматривается на основе степени засорения конструктивных элементов. Измерение процесса старения компонентов осуществляется в определенные промежутки времени. Благодаря такому подходу данные, полученные в неравномерно распределенные промежутки времени, предварительно обрабатываются для формирования сравниваемых бортовых показателей. На первом этапе агрегируем данные в определенные интервалы. Тем самым динамическая база бортовых данных уменьшается, что позволяет более эффективно анализировать сигналы. Также используем алгоритм машинного обучения с целью создания цифровой модели для измерения процесса старения. С помощью методологии локальных интерпретируемых модельно-агностических объяснений модель становится интерпретируемой. Это позволяет извлекать наиболее релевантные сигналы и тем самым сокращать объем обрабатываемых данных. Полученные результаты показывают, что для прогнозирования процесса старения рассматриваемого структурного компонента достаточно определенного количества бортовых сигналов. Таким образом, предлагаемый подход позволяет сократить передачу данных бортовых сигналов для проведения профилактического обслуживания.

Predictive maintenance has become important for avoiding unplanned downtime of modern vehicles. With increasing functionality the exchanged data between Electronic Control Units (ECU) grows simultaneously rapidly. A large number of in-vehicle signals are provided for monitoring an aging process. Various components of a vehicle age due to their usage. This component aging is only visible in a certain number of in-vehicle signals. In this work, we present a signal selection method for in-vehicle signals in order to determine relevant signals to monitor and predict powertrain component aging of vehicles. Our application considers the aging of powertrain components with respect to clogging of structural components. We measure the component aging process in certain time intervals. Owing to this, unevenly spaced time series data is preprocessed to generate comparable in-vehicle data. First, we aggregate the data in certain intervals. Thus, the dynamic in-vehicle database is reduced which enables us to analyze the signals more efficiently. Secondly, we implement machine learning algorithms to generate a digital model of the measured aging process. With the help of Local Interpretable Model-Agnostic Explanations (LIME) the model gets interpretable. This allows us to extract the most relevant signals and to reduce the amount of processed data. Our results show that a certain number of in-vehicle signals are sufficient for predicting the aging process of the considered structural component. Consequently, our approach allows to reduce data transmission of in-vehicle signals with the goal of predictive maintenance.

прогнозное техническое обслуживаниевыделение признаковвыбор сигналавременные рядымашинное обучениеобъяснение моделиpredictive maintenancefeature extractionsignal selectiontime seriesmachine learningmodel explanationReferencesGoyal D, Pabla B. S. (2015) Condition based maintenance of machine tools–A review. CIRP Journal of Manufacturing Science and Technology, 10, 24–35. https://doi.org/10.1016/j.cirpj.2015.05.004Goyal D, Pabla B. S. (2015) Condition based maintenance of machine tools–A review. CIRP Journal of Manufacturing Science and Technology, 10, 24–35. https://doi.org/10.1016/j.cirpj.2015.05.004Teti R, Jemielniak K, O’Donnell G, Dornfeld D. (2010) Advanced monitoring of machining operations. CIRP Annals, 59 (2), 717–739. https://doi.org/10.1016/j.cirp.2010.05.010Teti R, Jemielniak K, O’Donnell G, Dornfeld D. (2010) Advanced monitoring of machining operations. CIRP Annals, 59 (2), 717–739. https://doi.org/10.1016/j.cirp.2010.05.010Bediaga I, Mendizabal X, Arnaiz A, Munoa J (2013) Ball bearing damage detection using traditional signal processing algorithms. IEEE Instrumentation & Measurement Magazine, 16 (2), 20–25. https://doi.org/10.1109/mim.2013.6495676Bediaga I, Mendizabal X, Arnaiz A, Munoa J (2013) Ball bearing damage detection using traditional signal processing algorithms. IEEE Instrumentation & Measurement Magazine, 16 (2), 20–25. https://doi.org/10.1109/mim.2013.6495676Guo H, Crossman J. A., Murphey Y. L., Coleman M. (2000) Automotive signal diagnostics using wavelets and machine learning. IEEE transactions on vehicular technology, 49 (5), 1650–1662. https://doi.org/10.1109/25.892549Guo H, Crossman J. A., Murphey Y. L., Coleman M. (2000) Automotive signal diagnostics using wavelets and machine learning. IEEE transactions on vehicular technology, 49 (5), 1650–1662. https://doi.org/10.1109/25.892549Carino J. A, Delgado-Prieto M., Iglesias J. A., Sanchis A., Zurita D., Millan M., Ortega Redondo J. A., Romero-Troncoso R. (2018) Fault Detection and Identification Methodology Under an Incremental Learning Framework Applied to Industrial Machinery. IEEE Access, 6, 49755–49766. https://doi.org/10.1109/access.2018.2868430Carino J. A, Delgado-Prieto M., Iglesias J. A., Sanchis A., Zurita D., Millan M., Ortega Redondo J. A., Romero-Troncoso R. (2018) Fault Detection and Identification Methodology Under an Incremental Learning Framework Applied to Industrial Machinery. IEEE Access, 6, 49755–49766. https://doi.org/10.1109/access.2018.2868430Carino J. A., Delgado-Prieto M., Zurita D., Millan M., Ortega Redondo J. A., Romero-Troncoso R. (2016) Enhanced Industrial Machinery Condition Monitoring Methodology Based on Novelty Detection and Multi-Modal Analysis. IEEE Access, 4, 7594–7604. https://doi.org/10.1109/access.2016.2619382Carino J. A., Delgado-Prieto M., Zurita D., Millan M., Ortega Redondo J. A., Romero-Troncoso R. (2016) Enhanced Industrial Machinery Condition Monitoring Methodology Based on Novelty Detection and Multi-Modal Analysis. IEEE Access, 4, 7594–7604. https://doi.org/10.1109/access.2016.2619382Ladommatos N., Balian R., Horrocks R., Cooper L. (1996) The Effect of Exhaust Gas Recirculation on Combustion and NOx Emissions in a High-Speed Direct-injection Diesel Engine. SAE Technical Paper Series. https://doi.org/10.4271/960840Ladommatos N., Balian R., Horrocks R., Cooper L. (1996) The Effect of Exhaust Gas Recirculation on Combustion and NOx Emissions in a High-Speed Direct-injection Diesel Engine. SAE Technical Paper Series. https://doi.org/10.4271/960840Zelenka P., Aufinger H., Reczek W., Cartellieri W. (1998) Cooled EGR A Key Technology for Future Efficient HD Diesels. SAE Technical Paper Series. https://doi.org/10.4271/980190.Zelenka P., Aufinger H., Reczek W., Cartellieri W. (1998) Cooled EGR A Key Technology for Future Efficient HD Diesels. SAE Technical Paper Series. https://doi.org/10.4271/980190.Hoard J., Abarham M., Styles D., Giuliano J. M., Sluder C. S., Storey J.M.E. (2008) Diesel EGR Cooler Fouling. SAE International Journal of Engines, 1 (1), 1234–1250. https://doi.org/10.4271/2008-01-2475Hoard J., Abarham M., Styles D., Giuliano J. M., Sluder C. S., Storey J.M.E. (2008) Diesel EGR Cooler Fouling. SAE International Journal of Engines, 1 (1), 1234–1250. https://doi.org/10.4271/2008-01-2475Bravo Y., Moreno F., Longo O. (2007) Improved Characterization of Fouling in Cooled EGR Systems. SAE Technical Paper Series. https://doi.org/10.4271/2007-01-1257.Bravo Y., Moreno F., Longo O. (2007) Improved Characterization of Fouling in Cooled EGR Systems. SAE Technical Paper Series. https://doi.org/10.4271/2007-01-1257.Hui K. H., Ooi C. S., Lim M. H., Leong M. S., Al-Obaidi S. M. (2017) An improved wrapper-based feature selection method for machinery fault diagnosis. PLOS ONE, 12 (12), e0189143. https://doi.org/10.1371/journal.pone.0189143Hui K. H., Ooi C. S., Lim M. H., Leong M. S., Al-Obaidi S. M. (2017) An improved wrapper-based feature selection method for machinery fault diagnosis. PLOS ONE, 12 (12), e0189143. https://doi.org/10.1371/journal.pone.0189143Prytz R., Nowaczyk S., Byttner S. (2011) Towards relation discovery for diagnostics. Proceedings of the First International Workshop on Data Mining for Service and Maintenance KDD4Service ’11. ACM Press, San Diego, California, 23–27. https://doi.org/10.1145/2018673.2018678Prytz R., Nowaczyk S., Byttner S. (2011) Towards relation discovery for diagnostics. Proceedings of the First International Workshop on Data Mining for Service and Maintenance KDD4Service ’11. ACM Press, San Diego, California, 23–27. https://doi.org/10.1145/2018673.2018678Zhang B., Zhang L., Xu J. (2016) Degradation Feature Selection for Remaining Useful Life Prediction of Rolling Element Bearings. Quality and Reliability Engineering International, 32 (2), 547–554. https://doi.org/10.1002/qre.1771Zhang B., Zhang L., Xu J. (2016) Degradation Feature Selection for Remaining Useful Life Prediction of Rolling Element Bearings. Quality and Reliability Engineering International, 32 (2), 547–554. https://doi.org/10.1002/qre.1771Mrowca A., Moser B., Gunnemann S. (2018) Discovering Groups of Signals in In-Vehicle Network Traces for Redundancy Detection and Functional Grouping. Machine Learning and Knowledge Discovery in Databases, Springer, Cham, 86-102. https://doi.org/10.1007/978-3-030-10997-4_6Mrowca A., Moser B., Gunnemann S. (2018) Discovering Groups of Signals in In-Vehicle Network Traces for Redundancy Detection and Functional Grouping. Machine Learning and Knowledge Discovery in Databases, Springer, Cham, 86-102. https://doi.org/10.1007/978-3-030-10997-4_6Crossman J. A., Hong Guo, Murphey Y. L., Cardillo J. (2003) Automotive signal fault diagnostics. I. Signal fault analysis, signal segmentation, feature extraction and quasi-optimal feature selection. IEEE Transactions on Vehicular Technology, 52, 1063–1075. https://doi.org/10.1109/tvt.2002.807635Crossman J. A., Hong Guo, Murphey Y. L., Cardillo J. (2003) Automotive signal fault diagnostics. I. Signal fault analysis, signal segmentation, feature extraction and quasi-optimal feature selection. IEEE Transactions on Vehicular Technology, 52, 1063–1075. https://doi.org/10.1109/tvt.2002.807635Kane M. J., Price N., Scotch M., Rabinowitz P. (2014) Comparison of ARIMA and Random Forest time series models for prediction of avian influenza H5N1 outbreaks. BMC Bioinformatics, 15 (1). https://doi.org/10.1186/1471-2105-15-276.Kane M. J., Price N., Scotch M., Rabinowitz P. (2014) Comparison of ARIMA and Random Forest time series models for prediction of avian influenza H5N1 outbreaks. BMC Bioinformatics, 15 (1). https://doi.org/10.1186/1471-2105-15-276.Ribeiro M. T., Singh S., Guestrin C. (2016) “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining KDD ’16. ACM Press, San Francisco, California, USA, 1135–1144. https://doi.org/10.1145/2939672.2939778Ribeiro M. T., Singh S., Guestrin C. (2016) “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining KDD ’16. ACM Press, San Francisco, California, USA, 1135–1144. https://doi.org/10.1145/2939672.2939778Sass A. U., Esatbeyoglu E., Fischer T. (2019) Monitoring of Powertrain Component Aging Using In-Vehicle Signals. Diagnose in mechatronischen Fahrzeugsystemen XIII: Neue Verfahren für Test, Prüfung und Diagnose von E/E-Systemen im Kfz. Books on Demand, 15-28.Sass A. U., Esatbeyoglu E., Fischer T. (2019) Monitoring of Powertrain Component Aging Using In-Vehicle Signals. Diagnose in mechatronischen Fahrzeugsystemen XIII: Neue Verfahren für Test, Prüfung und Diagnose von E/E-Systemen im Kfz. Books on Demand, 15-28.

The authors declare that there are no conflicts of interest present.