Using machine learning and optimization for controlling surface roughness in grinding of St37

Dehghanpour Abyaneh, Mohsen; Narimani, Parviz; Hadad, Mohammadjafar; Attarsharghi, Samareh

doi:10.22059/ees.2023.2000614.1424

Using machine learning and optimization for controlling surface roughness in grinding of St37

Document Type : Research Paper

Authors

¹ School of Mechanical Engineering, College of Engineering, University of Tehran P.O. Box 11155/4563, Tehran, Iran

² School of Mechanical Engineering, College of Engineering, University of Tehran, Tehran, Iran

³ Department of Mechanical Engineering, School of Engineering and Technology, University of Doha for Science and Technology, Doha, Qatar

⁴ Department of Electrical Engineering, School of Engineering and Technology, University of Doha for Science and Technology, Doha, Qatar

10.22059/ees.2023.2000614.1424

Abstract

In the context of the industrial grinding process, the quality of products is often assessed by the final surface roughness, which is influenced by various parameters in the industrial environment. Previous studies lacked a feasible formulation based on a kinematical and statistical model to explain the uncertainty and non-linearity of grinding conditions, particularly concerning the cooling method, leading to significant discrepancies between the formulated and real results. This study introduces a novel strategy that combines deep learning and optimization to establish a suitable framework. It employs an artificial neural network to simulate and predict surface roughness, considering various dressing and cooling parameters in the industrial grinding of St37 steel alloy. Initially, an analysis of variance (ANOVA) is conducted to determine the correlation between input and output data. Subsequently, a neural network approach with one and two hidden layers, incorporating various activation functions, is employed. Therefore controlling and improving the accuracy of surface roughness predictions in industrial grinding processes can be automated. The mean squared error (MSE) metric is applied to each implementation to identify the best network architecture for the dataset. Upon selecting the network with the lowest MSE, the final algorithm predicts a set of randomly selected data from the dataset, achieving an overall accuracy of 80%. When compared to the accuracy of the formulated implementation, the neural network approach demonstrates a significantly higher accuracy of up to 30%, surpassing conventional analytical formulation in predicting final surface roughness. These results underscore the considerable potential and feasibility of deep learning approaches for industrial applications.

Keywords

References

[1] hare SK, Agarwal S (2015) Predictive Modeling of Surface Roughness in Grinding. Procedia CIRP 31:375–380. https://doi.org/10.1016/j.procir.2015.04.092

[2] Pan Y, Zhou P, Yan Y, et al (2021) New insights into the methods for predicting ground surface roughness in the age of digitalisation. Precis Eng 67:393–418. https://doi.org/10.1016/j.precisioneng.2020.11.001

[3] Palmer J, Ghadbeigi H, Novovic D, Curtis D (2018) an experimental study of the effects of dressing parameters on the topography of grinding wheels during roller dressing. J Manuf Process 31:348–355. https://doi.org/10.1016/j.jmapro.2017.11.025

[4] Zhou X, Xi F (2002) Modeling and predicting surface roughness of the grinding process. Int J Mach Tools Manuf 42:969–977. https://doi.org/10.1016/S0890-6955(02)00011-1

[5] Krishnan BR (2020) Review of surface roughness prediction in machining process by using various parameters. International Journal of Recent Trends in Engineering & Research (IJRTER) 6:7–12

[6] Gupta MK, Khan AM, Song Q, et al (2021) A review on conventional and advanced minimum quantity lubrication approaches on performance measures of grinding process. The International Journal of Advanced Manufacturing Technology 2021 117:3 117:729–750. https://doi.org/10.1007/S00170-021-07785-X

[7] Elbah M, Yallese MA, Aouici H, et al (2013) Comparative assessment of wiper and conventional ceramic tools on surface roughness in hard turning AISI 4140 steel. Measurement 46:3041–3056. https://doi.org/10.1016/J.MEASUREMENT.2013.06.018

[8] Jiang JL, Ge PQ, Bi WB, et al (2013) 2D/3D ground surface topography modeling considering dressing and wear effects in grinding process. Int J Mach Tools Manuf 74:29–40. https://doi.org/10.1016/J.IJMACHTOOLS.2013.07.002

[9] Liu Y, Warkentin A, Bauer R, Gong Y (2013) Investigation of different grain shapes and dressing to predict surface roughness in grinding using kinematic simulations. Precis Eng 37:758–764. https://doi.org/10.1016/J.PRECISIONENG.2013.02.009

[10] D’Addona DM, Matarazzo D, de Aguiar PR, et al (2016) Neural Networks Tool Condition Monitoring in Single-point Dressing Operations. Procedia CIRP 41:431–436. https://doi.org/10.1016/J.PROCIR.2016.01.001

[11] Sangwan KS, Saxena S, Kant G (2015) Optimization of Machining Parameters to Minimize Surface Roughness using Integrated ANN-GA Approach. Procedia CIRP 29:305–310. https://doi.org/10.1016/J.PROCIR.2015.02.002

[12] Moia DFG, Thomazella IH, Aguiar PR, et al (2014) Tool condition monitoring of aluminum oxide grinding wheel in dressing operation using acoustic emission and neural networks. Journal of the Brazilian Society of Mechanical Sciences and Engineering 2014 37:2 37:627–640. https://doi.org/10.1007/S40430-014-0191-6

[13] Holesovsky F, Pan B, Morgan MN, Czan A (2018) Evaluation of Diamond Dressing Effect on Workpiece Surface Roughness by Way of Analysis of Variance. Tehnički vjesnik 25:165–169. https://doi.org/10.17559/TV-20160411122230

[14] Hadad M, Sharbati A (2016) Analysis of the effects of dressing and wheel topography on grinding process under different coolant-lubricant conditions. The International Journal of Advanced Manufacturing Technology 2016 90:9 90:3727–3738. https://doi.org/10.1007/S00170-016-9703-0

[15] Masoudi S, Vafadar A, Hadad M, Jafarian F (2017) Experimental investigation into the effects of nozzle position, workpiece hardness, and tool type in MQL turning of AISI 1045 steel. https://doi.org/101080/1042691420171401716 33:1011–1019. https://doi.org/10.1080/10426914.2017.1401716

[16] Hadad MJ, Tawakoli T, Sadeghi MH, Sadeghi B (2012) Temperature and energy partition in minimum quantity lubrication-MQL grinding process. Int J Mach Tools Manuf 54–55:10–17. https://doi.org/10.1016/J.IJMACHTOOLS.2011.11.010

[17] Hadad M, Makarian J (2021) Experimental investigation of the effects of dressing and coolant-lubricant conditions on grinding of Nickel-based superalloy-Inconel 738. Energy Equipment and Systems 9:27–36. https://doi.org/10.22059/EES.2021.243054

[18] Chandrasekaran M, Devarasiddappa D (2014) Artificial neural network modeling for surface roughness prediction in cylindrical grinding of Al-SiCp metal matrix composites and ANOVA analysis. Advances in Production Engineering And Management 9:59–70. https://doi.org/10.14743/APEM2014.2.176

[19] Sahid N, … MR-IJ of, 2015 undefined NEURAL NETWORK MODELING OF GRINDING PARAMETERS OF DUCTILE CAST IRON USING MINIMUM QUANTITY LUBRICATION. pdfs.semanticscholar.org ISSN:2608–2621. https://doi.org/10.15282/ijame.11.2015.39.0220

[20] Govindhasamy JJ, McLoone SF, Irwin GW, et al (2005) Neural modelling, control and optimisation of an industrial grinding process. Control Eng Pract 13:1243–1258. https://doi.org/10.1016/J.CONENGPRAC.2004.11.006

[21] Prabhu S, Uma M, Vinayagam BK (2014) Surface roughness prediction using Taguchi-fuzzy logic-neural network analysis for CNT nanofluids based grinding process. Neural Computing and Applications 2014 26:1 26:41–55. https://doi.org/10.1007/S00521-014-1696-8

[22] Guo W, Wu C, Ding Z, Zhou Q (2021) Prediction of surface roughness based on a hybrid feature selection method and long short-term memory network in grinding. International Journal of Advanced Manufacturing Technology 112:2853–2871. https://doi.org/10.1007/S00170-020-06523-Z/TABLES/10

[23] Dijmărescu MR, Abaza BF, Voiculescu I, et al (2021) Surface Roughness Analysis and Prediction with an Artificial Neural Network Model for Dry Milling of Co–Cr Biomedical Alloys. Materials 2021, Vol 14, Page 6361 14:6361. https://doi.org/10.3390/MA14216361

[24] Zayegh A, systems NAB-D, 2018 undefined (2018) Neural network principles and applications

[25] Silaparasetty V (2020) Deep Learning Projects Using TensorFlow 2. Apress

[26] Michelucci U (2018) Applied deep learning: A case-based approach to understanding deep neural networks. Apress Media LLC

[27] Paluszek M, Thomas S (2020) Practical MATLAB deep learning: A project-based approach. Apress Media LLC

[28] Paliwal M, Kumar UA (2009) Neural networks and statistical techniques: A review of applications. Expert Syst Appl 36:2–17. https://doi.org/10.1016/J.ESWA.2007.10.005

[29] Pai Aravindpai (2020) Pai, A. (2020). CNN vs. RNN vs. ANN—analyzing 3... - Google Scholar. https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Pai%2C+A.+%282020%29.+CNN+vs.+RNN+vs.+ANN%E2%80%94analyzing+3+types+of+neural+networks+in+deep+learning.+Analytics+Vidhya%2C+Feb%2C+17.&btnG=. Accessed 17 Sep 2022

[30] Svozil D, Kvasnička V, Pospíchal J (1997) Introduction to multi-layer feed-forward neural networks. Chemometrics and Intelligent Laboratory Systems 39:43–62. https://doi.org/10.1016/S0169-7439(97)00061-0

[31] Rabiei F, Rahimi AR, Hadad MJ, Ashrafijou M (2015) Performance improvement of minimum quantity lubrication (MQL) technique in surface grinding by modeling and optimization. J Clean Prod 86:447–460. https://doi.org/10.1016/J.JCLEPRO.2014.08.045

[32] Gupta MK, Khan AM, Song Q, et al (2021) A review on conventional and advanced minimum quantity lubrication approaches on performance measures of grinding process. The International Journal of Advanced Manufacturing Technology 2021 117:3 117:729–750. https://doi.org/10.1007/S00170-021-07785-X

[33] Majumdar S, Das P, Kumar S, et al (2021) Evaluation of cutting fluid application in surface grinding. Measurement 169:108464. https://doi.org/10.1016/J.MEASUREMENT.2020.108464

[34] Aurich JC, Herzenstiel P, Sudermann H, Magg T (2008) High-performance dry grinding using a grinding wheel with a defined grain pattern. CIRP Annals 57:357–362. https://doi.org/10.1016/J.CIRP.2008.03.093

[35] Ni J, Yang Y, Wu C (2019) Assessment of water-based fluids with additives in grinding disc cutting process. J Clean Prod 212:593–601. https://doi.org/10.1016/J.JCLEPRO.2018.12.066

[36] Nadolny K, Kieraś S (2020) New approach for cooling and lubrication in dry machining on the example of internal cylindrical grinding of bearing rings. Sustainable Materials and Technologies 24:e00166. https://doi.org/10.1016/J.SUSMAT.2020.E00166

[37] Sarikaya M, Gupta MK, Tomaz I, et al (2021) Cooling techniques to improve the machinability and sustainability of light-weight alloys: A state-of-the-art review. J Manuf Process 62:179–201. https://doi.org/10.1016/J.JMAPRO.2020.12.013

[38] Guo C, Malkin S (2000) Energy Partition and Cooling During Grinding. J Manuf Process 2:151–157. https://doi.org/10.1016/S1526-6125(00)70116-2

[39] Irani RA, Bauer RJ, Warkentin A (2005) A review of cutting fluid application in the grinding process. Int J Mach Tools Manuf 45:1696–1705. https://doi.org/10.1016/J.IJMACHTOOLS.2005.03.006

[40] Klocke F, Thiermann J, Mattfeld P (2015) Influence of the dressing process on grinding wheel wear. Production Engineering 2015 9:5 9:563–568. https://doi.org/10.1007/S11740-015-0606-Y

[41] Deng H, Xu Z (2019) Dressing methods of superabrasive grinding wheels: A review. J Manuf Process 45:46–69. https://doi.org/10.1016/J.JMAPRO.2019.06.020

[42] Ding W, Li H, Zhang L, et al (2017) Diamond wheel dressing: A comprehensive review. Journal of Manufacturing Science and Engineering, Transactions of the ASME 139:. https://doi.org/10.1115/1.4037991/473220

[43] Bailly A, Blanc C, Francis É, et al (2022) Effects of dataset size and interactions on the prediction performance of logistic regression and deep learning models. Comput Methods Programs Biomed 213:106504. https://doi.org/10.1016/J.CMPB.2021.106504

[44] Brownlee J Impact of dataset size on deep learning model skill... - Google Scholar. https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Impact+of+dataset+size+on+deep+learning+model+skill+and+performance+estimates.+Machine+Learning+Mastery&btnG=. Accessed 18 Sep 2022

[45] McHugh ML (2011) Multiple comparison analysis testing in ANOVA. Biochem Med (Zagreb) 21:203–209

[46] Judd CM, McClelland GH, Ryan CS (2017) Data Analysis : A Model Comparison Approach to Regression, ANOVA, and Beyond. https://doi.org/10.4324/9781315744131

[47] Jierula A, Wang S, Oh TM, Wang P (2021) Study on Accuracy Metrics for Evaluating the Predictions of Damage Locations in Deep Piles Using Artificial Neural Networks with Acoustic Emission Data. Applied Sciences 2021, Vol 11, Page 2314 11:2314. https://doi.org/10.3390/APP11052314

[48] Kvalseth TO (2012) Cautionary Note about R 2. http://dx.doi.org/101080/00031305198510479448 39:279–285. https://doi.org/10.1080/00031305.1985.10479448

[49] Applied RH-FTIJ of, 2006 undefined Another look at forecast-accuracy metrics for intermittent demand. Citeseer