Iranian Journal of Manufacturing Engineering

Iranian Journal of Manufacturing Engineering

Developing surface roughness prediction model using deep convolutional network

Document Type : Original Article

Authors
Mohammad Zangene Najafi
Mohammad Shahbazi
Seyed Ali Niknam
School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran
10.22034/ijme.2024.474664.2000
Abstract
Surface roughness of machined parts is a critical parameter indicating surface quality, influenced by various factors. This report presents a deep learning-based framework for predicting and classifying surface roughness in milled parts. The model was developed using data from 162 experiments on three aluminum alloys: 7075, 6061, and 2024, which included cutting speed, cutting depth, tool coating type, feed rate, and surface roughness output. In this study, acoustic emission signals recorded during milling experiments were converted into two-dimensional images and fed into convolutional neural networks such as ResNet18, ShuffleNet, MobileNet and CNN-LSTM. Four encoding methods were used to convert time series signals into 2D images. The Segmented Stacked Permuted Channels (SSPC) method achieved the best performance with an accuracy above 98% across most models. Also, by increasing the number of classes, the accuracy values of the four mentioned methods have been investigated. ShuffleNet and MobileNet with an accuracy of 96-99% and low computational cost, is identified as suitable for real-time monitoring. The methods' efficiency in the mentioned networks was also evaluated under two noise levels (40% and 80%) in both zero-mean noise and non-zero-mean noise scenarios.
Keywords
Machining
Deep Learning
Surface Roughness
Acoustic Emission
[1] Serin G, Sener B, Ozbayoglu AM, Unver HO. Review of tool condition monitoring in machining and opportunities for deep learning. The International Journal of Advanced Manufacturing Technology. 2020 Jul;109(3):953-74. doi: 10.1007/s00170-020-05449-w
[2] Bai L, Yang Q, Cheng X, Ding Y, Xu J. A hybrid physics-data-driven surface roughness prediction model for ultra-precision machining. Science China Technological Sciences. 2023 May;66(5):1289-303. doi: 10.1007/s11431-022-2358-4
[3] Guo M, Wei S, Han C, Xia W, Luo C, Lin Z. Prediction of surface roughness using deep learning and data augmentation. Journal of Intelligent Manufacturing and Special Equipment. 2024 Jan 29. doi: 10.1108/JIMSE-10-2023-0010
[4] Lin WJ, Lo SH, Young HT, Hung CL. Evaluation of deep learning neural networks for surface roughness prediction using vibration signal analysis. Applied Sciences. 2019 Apr 8;9(7):1462. doi: 10.3390/app9071462
[5] Feng P, Borghesani P, Smith WA, Peng Z. Model-based surface roughness estimation using acoustic emission signals. Tribology International. 2020 Apr 1;144:106101. doi: 10.1016/j.triboint.2019.106101
[6] Abu-Mahfouz I, El Ariss O, Esfakur Rahman AH, Banerjee A. Surface roughness prediction as a classification problem using support vector machine. The International Journal of Advanced Manufacturing Technology. 2017 Sep;92:803-15. doi: 10.1007/s00170-017-0165-9
[7] Yeganefar A, Niknam SA, Asadi R. The use of support vector machine, neural network, and regression analysis to predict and optimize surface roughness and cutting forces in milling. The International Journal of Advanced Manufacturing Technology. 2019 Nov;105:951-65. doi: 10.1007/s00170-019-04227-7
[8] Dubey V, Sharma AK, Pimenov DY. Prediction of surface roughness using machine learning approach in MQL turning of AISI 304 steel by varying nanoparticle size in the cutting fluid. Lubricants. 2022 May 2;10(5):81. doi: 10.3390/lubricants10050081
[9] Kant G, Sangwan KS. Predictive modelling and optimization of machining parameters to minimize surface roughness using artificial neural network coupled with genetic algorithm. Procedia Cirp. 2015 Jan 1;31:453-8. doi: 10.1016/j.procir.2015.03.043
[10] Khorasani A, Yazdi MR. Development of a dynamic surface roughness monitoring system based on artificial neural networks (ANN) in milling operation. The International Journal of Advanced Manufacturing Technology. 2017 Oct;93:141-51. doi: 10.1007/s00170-015-7922-4
[11] Gupta AK, Guntuku SC, Desu RK, Balu A. Optimisation of turning parameters by integrating genetic algorithm with support vector regression and artificial neural networks. The International Journal of Advanced Manufacturing Technology. 2015 Mar;77:331-9. doi: 10.1007/s00170-014-6282-9
[12] Imani L, Rahmani Hanzaki A, Hamzeloo SR, Davoodi B. Modeling and optimization of cutting force and surface roughness in the milling process of Inconel 738 by Neural Network and Genetic Algorithm. Iranian Journal of Manufacturing Engineering. 2019 Oct 23;6(5):25-38. [In Persian]
[13] Siahvashi A, Shahbazi M, Niknam SA. The use of a neuro-fuzzy network coupled with meta-heuristic learning methods to predict surface roughness in the machining of aluminum alloys. Iranian Journal of Manufacturing Engineering. 2024;10(11):49-60. doi: 10.22034/ijme.2024.412021.1816 [In Persian]
[14] Huang PM, Lee CH. Estimation of tool wear and surface roughness development using deep learning and sensors fusion. Sensors. 2021 Aug 7;21(16):5338. doi: 10.3390/s21165338
[15] Pan Y, Kang R, Dong Z, Du W, Yin S, Bao Y. On-line prediction of ultrasonic elliptical vibration cutting surface roughness of tungsten heavy alloy based on deep learning. Journal of Intelligent Manufacturing. 2022 Mar 1:1-1. doi: 10.1007/s10845-020-01669-9
[16] Guo W, Wu C, Ding Z, Zhou Q. Prediction of surface roughness based on a hybrid feature selection method and long short-term memory network in grinding. The International Journal of Advanced Manufacturing Technology. 2021 Feb;112:2853-71. doi: 10.1007/s00170-020-06523-z
[17] Kirichenko L, Zinchenko P, Radivilova T. Classification of time realizations using machine learning recognition of recurrence plots. InInternational Scientific Conference “Intellectual Systems of Decision Making and Problem of Computational Intelligence” 2020 May 25 (pp. 687-696). Cham: Springer International Publishing. doi: 10.1007/978-3-030-54215-3_44