{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T15:21:36Z","timestamp":1777130496735,"version":"3.51.4"},"reference-count":61,"publisher":"MDPI AG","issue":"21","license":[{"start":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T00:00:00Z","timestamp":1761523200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"RCM2+"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Energies"],"abstract":"<jats:p>Data is an important resource for gaining knowledge about the behavior and condition monitoring of machines, enabling the estimation of parameters and the prediction of failures. However, in industrial environments, sensor interruptions often create gaps in the time series, which affects the reliability of the data. To overcome this challenge, this paper proposes an imputation strategy based on recurrent neural networks, in particular long short-term memory (LSTM) models, within a multivariate encoder\u2013decoder architecture. This approach utilizes correlations between variables to reconstruct missing values, resulting in more complete and robust datasets. Experimental results with multivariate time series show that the proposed method achieves accurate imputation, with errors as low as RMSE = 2.33 and R2 = 0.90 for some variables. Comparisons with alternative architectures, including GRU and Dense networks, show that LSTM excels in specific cases (e.g., VL3, R2 = 0.45), while the Dense architecture provides more stable performance across most variables. In particular, the Dense model achieved the best overall balance between accuracy and robustness, reaching RMSE = 2.33 and R2 = 0.90 for the best-performing variables, while the LSTM achieved the lowest error values in targeted scenarios, confirming its suitability for capturing complex temporal dependencies. Overall, this study highlights the feasibility of using recurrent neural networks to exploit temporal correlations for reliable data recovery, even under conditions of signal interruption in factory environments.<\/jats:p>","DOI":"10.3390\/en18215634","type":"journal-article","created":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T05:48:46Z","timestamp":1761716926000},"page":"5634","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Hybrid Deep Learning for Predictive Maintenance: LSTM, GRU, CNN, and Dense Models Applied to Transformer Failure Forecasting"],"prefix":"10.3390","volume":"18","author":[{"given":"Baldu\u00edno C\u00e9sar","family":"Mateus","sequence":"first","affiliation":[{"name":"RCM2+ Faculty of Engineering, Lus\u00f3fona University, 1749-024 Lisbon, Portugal"},{"name":"CISE\u2014Electromechatronic Systems Research Centre, University of Beira Interior, Cal\u00e7ada Fonte do Lameiro, 6201-001 Covilh\u00e3, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4313-7966","authenticated-orcid":false,"given":"Mateus","family":"Mendes","sequence":"additional","affiliation":[{"name":"Coimbra Institute of Engineering, Polytechnic University of Coimbra, Rua Pedro Nunes-Quinta da Nora, 3030-199 Coimbra, Portugal"},{"name":"RCM2+ Research Centre for Asset Management and Systems Engineering, Rua Pedro Nunes, 3030-199 Coimbra, Portugal"},{"name":"Department of Electrical and Computer Engineering, Institute of Systems and Robotics, University of Coimbra, 3030-290 Coimbra, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9694-8079","authenticated-orcid":false,"given":"Jos\u00e9 Torres","family":"Farinha","sequence":"additional","affiliation":[{"name":"Coimbra Institute of Engineering, Polytechnic University of Coimbra, Rua Pedro Nunes-Quinta da Nora, 3030-199 Coimbra, Portugal"},{"name":"RCM2+ Research Centre for Asset Management and Systems Engineering, Rua Pedro Nunes, 3030-199 Coimbra, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9947-0894","authenticated-orcid":false,"given":"Alexandre","family":"Martins","sequence":"additional","affiliation":[{"name":"RCM2+ Faculty of Engineering, Lus\u00f3fona University, 1749-024 Lisbon, Portugal"},{"name":"CISE\u2014Electromechatronic Systems Research Centre, University of Beira Interior, Cal\u00e7ada Fonte do Lameiro, 6201-001 Covilh\u00e3, Portugal"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,27]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Yazdi, M. (2024). Maintenance Strategies and Optimization Techniques. Advances in Computational Mathematics for Industrial System Reliability and Maintainability, Springer Nature.","DOI":"10.1007\/978-3-031-53514-7_3"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"6652","DOI":"10.1080\/00207543.2022.2126019","article-title":"Managing physical assets: A systematic review and a sustainable perspective","volume":"61","author":"Sandu","year":"2023","journal-title":"Int. J. Prod. Res."},{"key":"ref_3","unstructured":"Farinha, J.M.T. (2020). Asset Maintenance Engineering Methodologies, Taylor & Francis Ltd.. [1st ed.]."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Velmurugan, R.S., and Dhingra, T. (2021). Asset Maintenance in Operations-Intensive Organizations. Asset Maintenance Management in Industry: A Comprehensive Guide to Strategies, Practices and Benchmarking, Springer International Publishing.","DOI":"10.1007\/978-3-030-74154-9"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Farinha, J.T., Raposo, H.D.N., de-Almeida-e Pais, J.E., and Mendes, M. (2023). Physical Asset Life Cycle Evaluation Models\u2014A Comparative Analysis towards Sustainability. Sustainability, 15.","DOI":"10.20944\/preprints202309.1321.v1"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Mirshekali, H., Santos, A.Q., and Shaker, H.R. (2023). A Survey of Time-Series Prediction for Digitally Enabled Maintenance of Electrical Grids. Energies, 16.","DOI":"10.3390\/en16176332"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.neucom.2012.04.022","article-title":"Cascade BPN based transmission line overload prediction and preventive action by generation rescheduling","volume":"94","author":"Balaraman","year":"2012","journal-title":"Neurocomputing"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1016\/j.asoc.2007.05.001","article-title":"Prediction of transmission line overloading using intelligent technique","volume":"8","author":"Sharma","year":"2008","journal-title":"Appl. Soft Comput."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"110805","DOI":"10.1115\/1.4047856","article-title":"Intelligent Maintenance Systems and Predictive Manufacturing","volume":"142","author":"Lee","year":"2020","journal-title":"J. Manuf. Sci. Eng."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Sagharidooz, M., Soltanali, H., Farinha, J.T., Raposo, H.D.N., and de-Almeida-e Pais, J.E. (2024). Reliability, Availability, and Maintainability Assessment-Based Sustainability-Informed Maintenance Optimization in Power Transmission Networks. Sustainability, 16.","DOI":"10.3390\/su16156489"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Cachada, A., Barbosa, J., Leit\u00f1o, P., Gcraldcs, C.A., Deusdado, L., Costa, J., Teixeira, C., Teixeira, J., Moreira, A.H., and Moreira, P.M. (2018, January 4\u20137). Maintenance 4.0: Intelligent and Predictive Maintenance System Architecture. Proceedings of the 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation (ETFA), Torino, Italy.","DOI":"10.1109\/ETFA.2018.8502489"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"02008","DOI":"10.1051\/matecconf\/201712102008","article-title":"Application of IoT concept on predictive maintenance of industrial equipment","volume":"121","author":"Parpala","year":"2017","journal-title":"MATEC Web Conf."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"103335","DOI":"10.1016\/j.compind.2020.103335","article-title":"Maintenance transformation through Industry 4.0 technologies: A systematic literature review","volume":"123","author":"Silvestri","year":"2020","journal-title":"Comput. Ind."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1007\/s13198-013-0200-7","article-title":"Self-maintenance techniques: A smart approach towards self-maintenance system","volume":"5","author":"Singh","year":"2014","journal-title":"Int. J. Syst. Assur. Eng. Manag."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Mateus, B.C., Farinha, J.T., and Mendes, M. (2024). Fault Detection and Prediction for Power Transformers Using Fuzzy Logic and Neural Networks. Energies, 17.","DOI":"10.3390\/en17020296"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"111885","DOI":"10.1016\/j.asoc.2024.111885","article-title":"Prediction maintenance based on vibration analysis and deep learning\u2014A case study of a drying press supported on a Hidden Markov Model","volume":"163","author":"Martins","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"ref_17","unstructured":"McKinsey & Company (2025, May 23). A Smarter Way to Digitize Maintenance and Reliability. Available online: https:\/\/www.mckinsey.com\/capabilities\/operations\/our-insights\/a-smarter-way-to-digitize-maintenance-and-reliability."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Crespo Marquez, A., Gomez Fernandez, J.F., Mart\u00ednez-Gal\u00e1n Fern\u00e1ndez, P., and Guillen Lopez, A. (2020). Maintenance Management through Intelligent Asset Management Platforms (IAMP). Emerging Factors, Key Impact Areas and Data Models. Energies, 13.","DOI":"10.3390\/en13153762"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Zhang, X., Li, H., and Wu, P. (2022). Data-Driven Construction Safety Information Sharing System Based on Linked Data, Ontologies, and Knowledge Graph Technologies. Int. J. Environ. Res. Public Health, 19.","DOI":"10.3390\/ijerph19020794"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1080\/20464177.2024.2368914","article-title":"A machine learning-based data-driven method for risk analysis of marine accidents","volume":"24","author":"Feng","year":"2025","journal-title":"J. Mar. Eng. Technol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"103298","DOI":"10.1016\/j.compind.2020.103298","article-title":"Machine learning and reasoning for predictive maintenance in Industry 4.0: Current status and challenges","volume":"123","author":"Dalzochio","year":"2020","journal-title":"Comput. Ind."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"2039","DOI":"10.1109\/TII.2017.2670505","article-title":"A Manufacturing Big Data Solution for Active Preventive Maintenance","volume":"13","author":"Wan","year":"2017","journal-title":"IEEE Trans. Ind. Inform."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"107267","DOI":"10.1016\/j.cie.2021.107267","article-title":"Intelligent machine learning based total productive maintenance approach for achieving zero downtime in industrial machinery","volume":"157","author":"Devaraj","year":"2021","journal-title":"Comput. Ind. Eng."},{"key":"ref_24","unstructured":"(2007). Maintenance Terminology; Portuguese standard based on EN 13306:2007 (Standard No. NP EN 13306:2007)."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Paolanti, M., Romeo, L., Felicetti, A., Mancini, A., Frontoni, E., and Loncarski, J. (2018, January 2\u20134). Machine Learning approach for Predictive Maintenance in Industry 4.0. Proceedings of the 2018 14th IEEE\/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), Oulu, Finland.","DOI":"10.1109\/MESA.2018.8449150"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Malta, A., Farinha, T., and Mendes, M. (2023). Augmented Reality in Maintenance\u2014History and Perspectives. J. Imaging, 9.","DOI":"10.3390\/jimaging9070142"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"4585","DOI":"10.1109\/JIOT.2019.2957029","article-title":"Challenges to IoT-Enabled Predictive Maintenance for Industry 4.0","volume":"7","author":"Compare","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"100110","DOI":"10.1016\/j.sintl.2021.100110","article-title":"Significance of sensors for industry 4.0: Roles, capabilities, and applications","volume":"2","author":"Javaid","year":"2021","journal-title":"Sens. Int."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"320","DOI":"10.37394\/23203.2020.15.33","article-title":"Production optimization versus asset availability\u2014A review","volume":"15","author":"Mateus","year":"2020","journal-title":"WSEAS Trans. Syst. Control"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Elnahrawy, E., and Nath, B. (2003, January 19). Cleaning and querying noisy sensors. Proceedings of the 2nd ACM International Conference on Wireless Sensor Networks and Applications, San Diego, CA, USA. WSNA \u201903.","DOI":"10.1145\/941360.941362"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"McGrath, M.J., and Scanaill, C.N. (2013). Sensing and Sensor Fundamentals. Sensor Technologies: Healthcare, Wellness, and Environmental Applications, Apress.","DOI":"10.1007\/978-1-4302-6014-1"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Fraden, J. (2010). Sensor Characteristics. Handbook of Modern Sensors: Physics, Designs, and Applications, Springer.","DOI":"10.1007\/978-1-4419-6466-3_2"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1016","DOI":"10.1109\/PROC.1963.2383","article-title":"An Analysis of the factors which determine signal\/noise discrimination in pulsed-carrier systems","volume":"51","author":"North","year":"1963","journal-title":"Proc. IEEE"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Shuai, M., Xie, K., Chen, G., Ma, X., and Song, G. (2008, January 12\u201314). A Kalman Filter Based Approach for Outlier Detection in Sensor Networks. Proceedings of the 2008 International Conference on Computer Science and Software Engineering, Wuhan, China.","DOI":"10.1109\/CSSE.2008.1240"},{"key":"ref_35","unstructured":"Tan, Y.L., Sehgal, V., and Shahri, H.H. (2005). Sensoclean: Handling Noisy and Incomplete Data in Sensor Networks Using Modeling. [Ph.D. Thesis, University of Kaiserslautern]."},{"key":"ref_36","unstructured":"Zhai, X.B., Chen, B., and Zhu, K. (2019). Data Cleaning Based on Multi-sensor Spatiotemporal Correlation. Machine Learning and Intelligent Communications, Springer International Publishing."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1635","DOI":"10.1016\/j.compchemeng.2004.01.009","article-title":"On-line outlier detection and data cleaning","volume":"28","author":"Liu","year":"2004","journal-title":"Comput. Chem. Eng."},{"key":"ref_38","unstructured":"Hyndman, R.J., and Athanasopoulos, G. (2021). Forecasting: Principles and Practice, OTexts. [3rd ed.]."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Keogh, E., and Kasetty, S. (2003, January 24\u201327). On the need for time series data mining benchmarks: A survey and empirical demonstration. Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, Washington, DC, USA.","DOI":"10.1145\/775060.775062"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/2379776.2379788","article-title":"Time-series data mining","volume":"45","author":"Esling","year":"2012","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_41","first-page":"8050","article-title":"Semi-supervised learning with deep generative models for high-dimensional missing data imputation","volume":"33","author":"Yoon","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst. (NeurIPS)"},{"key":"ref_42","first-page":"1","article-title":"Missing value imputation for microarray gene expression data using a weighted KNN algorithm with truncated normal distribution","volume":"18","author":"Zhang","year":"2017","journal-title":"BMC Bioinform."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","article-title":"Maximum Likelihood from Incomplete Data via the EM Algorithm","volume":"39","author":"Dempster","year":"1977","journal-title":"J. R. Stat. Soc. Ser. B (Methodol.)"},{"key":"ref_44","unstructured":"Schmitt, M., and Zhu, X.X. (2019). Missing Data Imputation for Remote Sensing Time Series by Out-of-Sample EM. Machine Learning and Knowledge Discovery in Databases, Springer."},{"key":"ref_45","unstructured":"Wu, Z., Li, Z., and Liu, Q. (2023). Semi-supervised learning with missing not at random data. arXiv, Available online: https:\/\/arxiv.org\/abs\/2308.08872."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Hassanien, A., Rizk, R.Y., Pamucar, D., Darwish, A., and Chang, K.C. (2023, January 20\u201322). The Impact of Data Normalization on KNN Rendering. Proceedings of the 9th International Conference on Advanced Intelligent Systems and Informatics, Cairo, Egypt.","DOI":"10.1007\/978-3-031-43247-7"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"135","DOI":"10.11591\/csit.v4i2.pp135-142","article-title":"Investigating the impact of data scaling on the k-nearest neighbor algorithm","volume":"4","author":"Pagan","year":"2023","journal-title":"Comput. Sci. Inf. Technol."},{"key":"ref_48","first-page":"13","article-title":"Comparison of Min-Max normalization and Z-Score Normalization in the K-nearest neighbor (kNN) Algorithm to Test the Accuracy of Types of Breast Cancer","volume":"4","author":"Henderi","year":"2021","journal-title":"Int. J. Inform. Inf. Syst."},{"key":"ref_49","first-page":"29","article-title":"An investigation on disparity responds of machine learning algorithms to data normalization method","volume":"10","author":"Ahmed","year":"2022","journal-title":"Aro-Sci. J. Koya Univ."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"3639","DOI":"10.1007\/s10462-020-09934-2","article-title":"A review on fault detection and diagnosis techniques: Basics and beyond","volume":"54","author":"Abid","year":"2021","journal-title":"Artif. Intell. Rev."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Brito, L., Susto, G., Brito, J., and Duarte, M. (2022). An explainable artificial intelligence approach for fault detection. Signal Process. Syst.","DOI":"10.1016\/j.ymssp.2021.108105"},{"key":"ref_52","doi-asserted-by":"crossref","unstructured":"Shakiba, F., Azizi, S., Zhou, M., and Abusorrah, A. (2023). Application of machine learning methods in fault detection. Artif. Intell. Rev.","DOI":"10.1007\/s10462-022-10296-0"},{"key":"ref_53","unstructured":"Azamfar, M., Singh, J., Bravo-Imaz, I., and Lee, J. (2020). Multisensor data fusion for gearbox fault diagnosis. Signal Process. Syst."},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Kiranyaz, S., Ince, T., Abdeljaber, O., Avci, O., and Gabbouj, M. (2019). 1-D convolutional neural networks for signal processing applications. IEEE Trans. Signal Process.","DOI":"10.1109\/ICASSP.2019.8682194"},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Chen, Y., Rao, M., Feng, K., and Zuo, M. (2022). Physics-Informed LSTM hyperparameters selection for fault detection. Signal Process. Syst.","DOI":"10.1016\/j.ymssp.2022.108907"},{"key":"ref_56","unstructured":"Ali, M., Shabbir, M., and Liang, X. (2019). Machine learning-based fault diagnosis for synchronous motors. IEEE Trans."},{"key":"ref_57","unstructured":"Kumar, P., and Hati, A. (2021). Review on machine learning algorithms for fault detection. Arch. Comput. Methods Eng."},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Yu, J., and Zhang, Y. (2023). Challenges and opportunities of deep learning-based predictive maintenance. Neural Comput. Appl.","DOI":"10.1007\/s00521-022-08017-3"},{"key":"ref_59","unstructured":"Samanta, A., Chowdhuri, S., and Williamson, S. (2021). Machine learning-based fault detection in electric vehicles. Electronics."},{"key":"ref_60","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_61","first-page":"12034","article-title":"Signal processing techniques for condition monitoring of electrical machines","volume":"9","author":"Javaid","year":"2021","journal-title":"IEEE Access"}],"container-title":["Energies"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1996-1073\/18\/21\/5634\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T06:04:43Z","timestamp":1761717883000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1996-1073\/18\/21\/5634"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,27]]},"references-count":61,"journal-issue":{"issue":"21","published-online":{"date-parts":[[2025,11]]}},"alternative-id":["en18215634"],"URL":"https:\/\/doi.org\/10.3390\/en18215634","relation":{},"ISSN":["1996-1073"],"issn-type":[{"value":"1996-1073","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,27]]}}}