Abstract
Digital twins, virtual representations of real-life physical objects or processes, are becoming widely used in many different industrial sectors. One of the main uses of digital twins is predictive maintenance, and these technologies are being adapted to various new applications and datatypes in many industrial processes. The aim of this study was to propose a methodology to generate synthetic vibration data using a digital twin model and a predictive maintenance workflow, consisting of preprocessing, feature engineering, and classification model training, to classify faulty and healthy vibration data for state estimation. To assess the success of the proposed workflow, the mentioned steps were applied to a publicly available vibration dataset and the synthetic data from the digital twin, using five different state-of-the-art classification algorithms. For several of the classification algorithms, the accuracy result for the classification of healthy and faulty data achieved on the public dataset reached approximately 86%, and on the synthetic data, approximately 98%. These results showed the great potential for the proposed methodology, and future work in the area.
Highlights
This section contains a discussion of the results of the classification accuracy after model training for the Case Western Reserve University (CWRU) dataset and the synthetic data
The better-than-chance training accuracy for each algorithm, and the models’ ability to estimate the current state of the system with over 70% accuracy for some algorithms, showed that the algorithm works as desired for the CWRU dataset
The main goal of this study was to suggest a viable workflow to generate healthy and faulty synthetic vibration data, and to bring forward a feasible predictive maintenance algorithm to classify these faulty and healthy data. This main objective was met by a process involving creating an algorithm consisting of a digital twin model of the system, generating the synthetic data, preprocessing the synthetic data, and extracting condition indicators from the preprocessed data and machine learning model training
Summary
The main goal of PdM is to optimize the maintenance schedule by predicting failures in machineries and processes. Such a process will result in reductions in unplanned downtimes of machinery, and fatal breakdowns [1,2]. PdM reduces unnecessary maintenance, which further increases the machine’s life, considering that every maintenance operation causes downtime [3]. Another advantage of PdM is cost minimization, including minimizing fatal breakdowns and reducing the replacement of key components, which is closely related to the previously mentioned benefits [4]. Predictive maintenance may reduce maintenance costs by 25–35%, eliminate downtimes by 70–75%, reduce downtime by 35–45%, and increase productivity by 25–35% [1,2]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
