Abstract
In modern solutions of machine tools and technological devices for precision machining, rolling guides are used. Their advantages are low resistance to movement and the possibility of obtaining very good repeatability of positioning of working units. The main structural element is ball screws. During their utilization, there is gradual wear of the cooperating elements of the ball screw-nut assembly and deterioration of their functional properties. As the work is performed by the screw, its working surfaces are subject to abrasive wear, which results in a gradual loss of the accuracy of the machine and the technological quality of the manufactured products. It can also lead to failure due to damage to the ball screws. Applied preventive actions come down mainly to periodic inspections and maintenance of mechanisms. In order to prevent excessive wear of the rolling guides or even machine failure, the authors of the article proposed a prognostic model to determine the wear intensity of the ball screw based on backlash measurements in the lead screw-nut assembly. The article presents a developed model for the prediction of the wear intensity of the rolling guides based on artificial neural networks. The authors obtained a model of prediction accuracy of 81%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Macukow, B.: Neural Networks History of Research and Basic Models. Warsaw University of Technology, Warsaw (2011)
Kosiński, R.A.: Artificial Neural Networks. Scientific and Technical Publishing House (2007)
Kane, A., Kore, A., Khandale, A., Nigade, A., Joshi P.P.: Predictive maintenance using machine learning. Comput. Sci. (2022)
Salunkhe, T., Jamadar, N.I., Kivade, S.B.: Prediction of remaining useful life of mechanical components – a review 3, 6 (2018)
Abbasi, J.A.: Predictive maintenance in industrial machinery using machine learning. Master Programme in Data Science. Luleå University of Technology, Luleå (2021)
Raptodimos, Z., Lazakis, I.: An artificial neural network approach for predicting the performance of ship machinery equipment. In: Conference: International Conference on Maritime Safety and Operations, Glasgow, UK (2016)
Rivas, A., Fraile, J., Chamoso, P., González-Briones, A., Sittón, I., Corchado, J.: A predictive maintenance model using recurrent neural networks. In: Advances in Intelligent Systems and Computing, vol. 950, Springer International Publishing, Berlin/Heidelberg, Germany (2019)
Ilyas, A., Park, S.M., Engstrom, L., Leclerc, G., Madry, A.: Datamodels: Predicting Predictions from Training Data, Baltimore. In: Proceedings of the 39 th International Conference on Machine Learning 162, Maryland, USA (2022)
Picard, A.M., Vigouroux, D., Zamolodtchikov, P.: Leveraging influence functions for dataset exploration and cleaning. In: 11th European Congress Embedded Real Time Systems (ERTS 2022), Toulouse, France (2022)
Hara, S., Atsushi, N.I., Takanori, M.: Data cleansing for models trained with SGD. In: Part of Advances in Neural Information Processing Systems (NeurIPS), vol. 32 (2019)
Kenji, S., Yoshiyuki, K., Takuya, N.: Data cleansing for deep neural networks with storage-efficient approximation of influence functions. Journal arXiv preprint arXiv:2103.11807 (2021)
Brophy, J.: Exit through the training data: a look into instance-attribution explanations and efficient data deletion in machine learning. Technical report (2020)
Tadeusiewicz, R., Gąciarz, T., Borowik, B., Leper, B.: Discovering the properties of neural networks with the use of C#. PAU publications (2007)
Nowicki, K., Sempruch, J.: Problems Related to the Training of Artificial Intelligence of the Neural Network used to Model Fatigue Characteristics. UTP publications (2007)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Krajewska-Śpiewak, J., Serwatka, Ł., Gawlik, J. (2024). Prediction of the Wear Intensity of Rolling Guides with the Use of a Neural Network. In: Burduk, A., Batako, A.D.L., Machado, J., Wyczółkowski, R., Dostatni, E., Rojek, I. (eds) Intelligent Systems in Production Engineering and Maintenance III. ISPEM 2023. Lecture Notes in Mechanical Engineering. Springer, Cham. https://doi.org/10.1007/978-3-031-44282-7_7
Download citation
DOI: https://doi.org/10.1007/978-3-031-44282-7_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44281-0
Online ISBN: 978-3-031-44282-7
eBook Packages: EngineeringEngineering (R0)