{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,27]],"date-time":"2026-06-27T15:04:08Z","timestamp":1782572648291,"version":"3.54.5"},"reference-count":42,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2023,5,22]],"date-time":"2023-05-22T00:00:00Z","timestamp":1684713600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"NUAA","award":["xcxjh20211607"],"award-info":[{"award-number":["xcxjh20211607"]}]},{"name":"NUAA","award":["2021ZD0113103"],"award-info":[{"award-number":["2021ZD0113103"]}]},{"name":"New Generation Artificial Intelligence","award":["xcxjh20211607"],"award-info":[{"award-number":["xcxjh20211607"]}]},{"name":"New Generation Artificial Intelligence","award":["2021ZD0113103"],"award-info":[{"award-number":["2021ZD0113103"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Surface roughness is a key indicator of the quality of mechanical products, which can precisely portray the fatigue strength, wear resistance, surface hardness and other properties of the products. The convergence of current machine-learning-based surface roughness prediction methods to local minima may lead to poor model generalization or results that violate existing physical laws. Therefore, this paper combined physical knowledge with deep learning to propose a physics-informed deep learning method (PIDL) for milling surface roughness predictions under the constraints of physical laws. This method introduced physical knowledge in the input phase and training phase of deep learning. Data augmentation was performed on the limited experimental data by constructing surface roughness mechanism models with tolerable accuracy prior to training. In the training, a physically guided loss function was constructed to guide the training process of the model with physical knowledge. Considering the excellent feature extraction capability of convolutional neural networks (CNNs) and gated recurrent units (GRUs) in the spatial and temporal scales, a CNN\u2013GRU model was adopted as the main model for milling surface roughness predictions. Meanwhile, a bi-directional gated recurrent unit and a multi-headed self-attentive mechanism were introduced to enhance data correlation. In this paper, surface roughness prediction experiments were conducted on the open-source datasets S45C and GAMHE 5.0. In comparison with the results of state-of-the-art methods, the proposed model has the highest prediction accuracy on both datasets, and the mean absolute percentage error on the test set was reduced by 3.029% on average compared to the best comparison method. Physical-model-guided machine learning prediction methods may be a future pathway for machine learning evolution.<\/jats:p>","DOI":"10.3390\/s23104969","type":"journal-article","created":{"date-parts":[[2023,5,23]],"date-time":"2023-05-23T02:02:27Z","timestamp":1684807347000},"page":"4969","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":38,"title":["Milling Surface Roughness Prediction Based on Physics-Informed Machine Learning"],"prefix":"10.3390","volume":"23","author":[{"given":"Shi","family":"Zeng","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dechang","family":"Pi","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2023,5,22]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Pimenov, D.Y., Hassui, A., Wojciechowski, S., Mia, M., Magri, A., Suyama, D.I., Bustillo, A., Krolczyk, G., and Gupta, M.K. (2019). Effect of the Relative Position of the Face Milling Tool towards the Workpiece on Machined Surface Roughness and Milling Dynamics. Appl. Sci., 9.","DOI":"10.3390\/app9050842"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1515\/secm-2013-0143","article-title":"Investigation of experimental study of end milling of CFRP composite","volume":"22","author":"Kilickap","year":"2015","journal-title":"Sci. Eng. Compos. Mater."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"103687","DOI":"10.1016\/j.ijmachtools.2020.103687","article-title":"Surface integrity in metal machining\u2013Part I: Fundamentals of surface characteristics and formation mechanisms","volume":"162","author":"Liao","year":"2021","journal-title":"Int. J. Mach. Tools Manuf."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1007\/s12541-019-00247-1","article-title":"Investigation on micro-milling of Ti\u20136Al\u20134V alloy by PCD slotting-tools","volume":"21","author":"Han","year":"2020","journal-title":"Int. J. Precis. Eng. Manuf."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"6743","DOI":"10.1007\/s00170-022-08692-5","article-title":"Investigation on self-fabricated PCD cutter and its application in deep-and-narrow micro-grooves","volume":"119","author":"Han","year":"2022","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"103907","DOI":"10.1016\/j.ijmachtools.2022.103907","article-title":"Origin of the lateral return error in a five-axis ultraprecision machine tool and its influence on ball-end milling surface roughness","volume":"178","author":"Cui","year":"2022","journal-title":"Int. J. Mach. Tools Manuf."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1016\/j.jmatprotec.2006.11.029","article-title":"Surface roughness optimization in an end-milling operation using the Taguchi design method","volume":"184","author":"Zhang","year":"2007","journal-title":"J. Mater. Process. Technol."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"107860","DOI":"10.1016\/j.measurement.2020.107860","article-title":"Evaluation of turned and milled surfaces roughness using convolutional neural network","volume":"161","author":"Rifai","year":"2020","journal-title":"Measurement"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1016\/S0924-0136(01)00959-1","article-title":"Surface roughness prediction based on cutting parameters and tool vibrations in turning operations","volume":"118","author":"Abouelatta","year":"2001","journal-title":"J. Mater. Process. Technol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"118475","DOI":"10.1016\/j.conbuildmat.2020.118475","article-title":"A novel support vector regression (SVR) model for the prediction of splice strength of the unconfined beam specimens","volume":"248","author":"Ahmad","year":"2020","journal-title":"Constr. Build. Mater."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.procir.2020.01.034","article-title":"Automated quality assurance as an intelligent cloud service using machine learning","volume":"86","author":"Schreiber","year":"2019","journal-title":"Procedia CIRP"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1016\/j.polymertesting.2016.05.022","article-title":"Detecting surface roughness on SLS parts with various measuring techniques","volume":"53","author":"Launhardt","year":"2016","journal-title":"Polym. Test."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"977","DOI":"10.1088\/0022-3735\/22\/12\/001","article-title":"Measurement of surface roughness by a machine vision system","volume":"22","author":"Luk","year":"1989","journal-title":"J. Phys. E Sci. Instrum."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"106797","DOI":"10.1016\/j.ijmecsci.2021.106797","article-title":"Surface topography prediction in peripheral milling of thin-walled parts considering cutting vibration and material removal effect","volume":"211","author":"Zhuo","year":"2021","journal-title":"Int. J. Mech. Sci."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2755","DOI":"10.1007\/s00170-020-05599-x","article-title":"A predictive model of milling surface roughness","volume":"108","author":"Wang","year":"2020","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"3289","DOI":"10.1007\/s00170-021-07410-x","article-title":"An analytical prediction model of surface topography generated in 4-axis milling process","volume":"115","author":"Zhou","year":"2021","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1193","DOI":"10.1007\/s00170-022-10709-y","article-title":"The development of an ANN surface roughness prediction system of multiple materials in CNC turning","volume":"125","author":"Huang","year":"2023","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"101470","DOI":"10.1016\/j.aei.2021.101470","article-title":"Digital twin-driven surface roughness prediction and process parameter adaptive optimization","volume":"51","author":"Liu","year":"2022","journal-title":"Adv. Eng. Inform."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1300","DOI":"10.1016\/j.matpr.2022.04.126","article-title":"Surface roughness prediction in milling using long-short term memory modelling","volume":"64","author":"Manjunath","year":"2022","journal-title":"Mater. Today Proc."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"5807","DOI":"10.1007\/s00170-021-08535-9","article-title":"Prediction of micro milling force and surface roughness considering size-dependent vibration of micro-end mill","volume":"119","author":"Du","year":"2022","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Wang, Y., Wang, Y., Zheng, L., and Zhou, J. (2022). Online Surface Roughness Prediction for Assembly Interfaces of Vertical Tail Integrating Tool Wear under Variable Cutting Parameters. Sensors, 22.","DOI":"10.3390\/s22051991"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Wang, Z., Li, H., and Yu, T. (2022). Study on Surface Integrity and Surface Roughness Model of Titanium Alloy TC21 Milling Considering Tool Vibration. Appl. Sci., 12.","DOI":"10.3390\/app12084041"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Xu, J., Yan, F., Wan, X., Li, Y., and Zhu, Q. (2023). Surface Topography Model of Ultra-High Strength Steel AF1410 Based on Dynamic Characteristics of Milling System. Processes, 11.","DOI":"10.3390\/pr11020641"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1095","DOI":"10.1007\/s00170-022-10587-4","article-title":"Modelling and prediction of surface topography on machined slot side wall with single-pass end milling","volume":"124","author":"Lyu","year":"2023","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"107128","DOI":"10.1016\/j.engfailanal.2023.107128","article-title":"New neural network-based algorithm for predicting fatigue life of aluminum alloys in terms of machining parameters","volume":"146","author":"Kashyzadeh","year":"2023","journal-title":"Eng. Fail. Anal."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Chen, C.-H., Jeng, S.-Y., and Lin, C.-J. (2022). Prediction and Analysis of the Surface Roughness in CNC End Milling Using Neural Networks. Appl. Sci., 12.","DOI":"10.3390\/app12010393"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Dubey, V., Sharma, A.K., and Pimenov, D.Y. (2022). Prediction of Surface Roughness Using Machine Learning Approach in MQL Turning of AISI 304 Steel by Varying Nanoparticle Size in the Cutting Fluid. Lubricants, 10.","DOI":"10.3390\/lubricants10050081"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"4065","DOI":"10.1007\/s00170-022-10470-2","article-title":"A physics-informed machine learning model for surface roughness prediction in milling operations","volume":"123","author":"Wu","year":"2022","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_29","unstructured":"Li, Y., Yu, R., Shahabi, C., and Liu, Y. (2017). Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., and Zhang, W. (2021, January 2\u20139). Informer: Beyond efficient transformer for long sequence time-series forecasting. Proceedings of the AAAI Conference on Artificial Intelligence, Online.","DOI":"10.1609\/aaai.v35i12.17325"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Wu, Z., Pan, S., Long, G., Jiang, J., Chang, X., and Zhang, C. (2020, January 6\u201310). Connecting the dots: Multivariate time series forecasting with graph neural networks. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Online.","DOI":"10.1145\/3394486.3403118"},{"key":"ref_32","unstructured":"Guen, V.L., and Thome, N. (2020, January 13\u201319). Disentangling physical dynamics from unknown factors for unsupervised video prediction. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA."},{"key":"ref_33","unstructured":"He, X., Cao, H.L., and Zhu, B. (2020). AdvectiveNet: An Eulerian-Lagrangian Fluidic Reservoir for Point Cloud Processing. arXiv."},{"key":"ref_34","unstructured":"Seo, S., Meng, C., and Liu, Y. (2020, January 30). Physics-aware difference graph networks for sparsely-observed dynamics. Proceedings of the International Conference on Learning Representations, Addis Ababa, Ethiopia."},{"key":"ref_35","unstructured":"Iakovlev, V., Heinonen, M., and L\u00e4hdesm\u00e4ki, H. (2020). Learning continuous-time PDEs from sparse data with graph neural networks. arXiv."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"105061","DOI":"10.1016\/j.ijmecsci.2019.105061","article-title":"Modelling and analysis of surface topography generated in face milling operations","volume":"163","author":"Arizmendi","year":"2019","journal-title":"Int. J. Mech. Sci."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1016\/j.ijmachtools.2018.02.001","article-title":"Influencing factors and theoretical modeling methods of surface roughness in turning process: State-of-the-art","volume":"129","author":"He","year":"2018","journal-title":"Int. J. Mach. Tools Manuf."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"305","DOI":"10.1007\/s00170-018-3176-2","article-title":"Prediction of surface roughness in milling process using vibration signal analysis and artificial neural network","volume":"102","author":"Wu","year":"2019","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Mahjoub, S., Chrifi-Alaoui, L., Marhic, B., and Delahoche, L. (2022). Predicting Energy Consumption Using LSTM, Multi-Layer GRU and Drop-GRU Neural Networks. Sensors, 22.","DOI":"10.3390\/s22114062"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Liu, Z., Li, W., Feng, J., and Zhang, J. (2022). Research on Satellite Network Traffic Prediction Based on Improved GRU Neural Network. Sensors, 22.","DOI":"10.3390\/s22228678"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Choudhury, N.A., and Soni, B. (2023). An Adaptive Batch Size based-CNN-LSTM Framework for Human Activity Recognition in Uncontrolled Environment. IEEE Trans. Ind. Inform., 1\u20139.","DOI":"10.1109\/TII.2022.3229522"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"111893","DOI":"10.1016\/j.measurement.2022.111893","article-title":"A remaining life prediction of rolling element bearings based on a bidirectional gate recurrent unit and convolution neural network","volume":"202","author":"Shang","year":"2022","journal-title":"Measurement"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/10\/4969\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T19:39:59Z","timestamp":1760125199000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/10\/4969"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,22]]},"references-count":42,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2023,5]]}},"alternative-id":["s23104969"],"URL":"https:\/\/doi.org\/10.3390\/s23104969","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,5,22]]}}}