Improving the accuracy of machine-learning models with data from machine test repetitions

The modelling of machining processes by means of machine-learning algorithms is still based on principles that are especially adapted to mechanical approaches, in which very few inputs are varied with little repetition of experimental conditions. These principles might not be ideal to achieve accurate machine-learning models and they are certainly not aligned with the practicalities of industrial machining in factories. In this research the effect of a new strategy to improve machine-learning model accuracy is studied: experimental repetition. Tool-life prediction in the face-turning operations of AISI 1045 steel discs, depending on different cooling systems and tool geometries, is selected as a case study. Both the side rake and the relief angles of HSS tools are optimized using the Brandsma facing test under dry, MQL, and flooding conditions. Different machine-learning algorithms, such as regression trees, kNNs, artificial neural networks, and ensembles (bagging and Random Forest) are tested. On the one hand, the results of the study showed that artificial neural networks of Radial Basis Functions presented the highest model accuracy (11.4 mm RMSE), but required a very sensitive and complex tuning process. On the other hand, they demonstrated that ensembles, especially Random Forest, provided models with accuracy in the same range, but with no tuning procedure (12.8 mm RMSE). Secondly, the effect of an increased dataset size, by means of experimental repetition, is evaluated and compared with traditional experimental modelling that used average values. The results showed that some machine-learning techniques, including both ensemble types, significantly improved their accuracy with this strategy, by up to 23%. The results therefore suggested that the use of raw experimental data, rather than their averaged values, can achieve machine-learning models of higher accuracy for tool-wear processes.

[1]  A. E. Diniz,et al.  Gym Wear Set Women Sport Custom Fitness Wear High Waist Leggings Bra Clothing Fitness Sexy Yoga Wear Set Clothes , 2001 .

[2]  J. Paulo Davim,et al.  Machining : fundamentals and recent advances , 2008 .

[3]  M. Nalbant,et al.  The experimental investigation of the effects of uncoated, PVD- and CVD-coated cemented carbide inserts and cutting parameters on surface roughness in CNC turning and its prediction using artificial neural networks , 2009 .

[4]  Dong-Won Kim,et al.  Fuzzy logic based tool condition monitoring for end-milling , 2017 .

[5]  Antonio J. Sánchez Egea,et al.  MÁQUINAS, PROCESOS, PERSONAS Y DATOS, LAS CLAVES PARA LA REVOLUCIÓN 4.0 , 2018 .

[6]  Jiayun Wang,et al.  A fault diagnosis model based on weighted extension neural network for turbo-generator sets on small samples with noise , 2020 .

[7]  Surjya K. Pal,et al.  FUZZY RADIAL BASIS FUNCTION (FRBF) NETWORK BASED TOOL CONDITION MONITORING SYSTEM USING VIBRATION SIGNALS , 2010 .

[8]  Joaquim Ciurana,et al.  Improvement of surface roughness models for face milling operations through dimensionality reduction , 2012, Integr. Comput. Aided Eng..

[9]  Khamdi Mubarok,et al.  Smart manufacturing systems for Industry 4.0: Conceptual framework, scenarios, and future perspectives , 2018, Frontiers of Mechanical Engineering.

[10]  W. Zong,et al.  Modeling and simulation on the effect of tool rake angle in diamond turning of KDP crystal , 2019, Journal of Materials Processing Technology.

[11]  Aman Jantan,et al.  State-of-the-art in artificial neural network applications: A survey , 2018, Heliyon.

[12]  Connor Jennings,et al.  A Comparative Study on Machine Learning Algorithms for Smart Manufacturing: Tool Wear Prediction Using Random Forests , 2017 .

[13]  Jin Jiang,et al.  Erratum to: State-of-the-art methods and results in tool condition monitoring: a review , 2005 .

[14]  G. Budzik,et al.  Ecological trends in machining as a key factor in sustainable production – A review , 2019, Journal of Cleaner Production.

[15]  R. Suresh,et al.  Some studies on hard turning of AISI 4340 steel using multilayer coated carbide tool , 2012 .

[16]  Andres Bustillo,et al.  A regression-tree multilayer-perceptron hybrid strategy for the prediction of ore crushing-plate lifetimes , 2019, Journal of advanced research.

[17]  Vishal S. Sharma,et al.  Effect of tool geometry variation on finish turning - A Review , 2011 .

[18]  Álvar Arnaiz-González,et al.  Using artificial neural networks for the prediction of dimensional error on inclined surfaces manufactured by ball-end milling , 2015, The International Journal of Advanced Manufacturing Technology.

[19]  I. A. Kattan,et al.  Developing new trends of cutting tool geometry , 1996 .

[20]  Ercan Öztemel,et al.  Literature review of Industry 4.0 and related technologies , 2018, J. Intell. Manuf..

[21]  Guofeng Wang,et al.  Force Sensor Based Tool Condition Monitoring Using a Heterogeneous Ensemble Learning Model , 2014, Sensors.

[22]  Bernhard Sick,et al.  ON-LINE AND INDIRECT TOOL WEAR MONITORING IN TURNING WITH ARTIFICIAL NEURAL NETWORKS: A REVIEW OF MORE THAN A DECADE OF RESEARCH , 2002 .

[23]  Suleyman Yaldiz,et al.  Investigation of the effect of rake angle and approaching angle on main cutting force and tool tip temperature , 2006 .

[24]  M. Matuszewski,et al.  Using artificial intelligence models for the prediction of surface wear based on surface isotropy levels , 2018, Robotics and Computer-Integrated Manufacturing.

[25]  Andrés Bustillo,et al.  Interpreting tree-based prediction models and their data in machining processes , 2016, Integr. Comput. Aided Eng..

[26]  J. A. Leonard,et al.  Radial basis function networks for classifying process faults , 1991, IEEE Control Systems.

[27]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[28]  Ron Kohavi,et al.  The Power of Decision Tables , 1995, ECML.

[29]  Andres Bustillo,et al.  Machine-learning for automatic prediction of flatness deviation considering the wear of the face mill teeth , 2020, Journal of Intelligent Manufacturing.

[30]  Jun Zhao,et al.  Effect of cutting parameters and tool rake angle on the chip formation and adiabatic shear characteristics in machining Ti-6Al-4V titanium alloy , 2020 .

[31]  P. V. Rao,et al.  A surface roughness prediction model for hard turning process , 2007 .

[32]  George-Christopher Vosniakos,et al.  Prediction of surface roughness in CNC face milling using neural networks and Taguchi's design of experiments , 2002 .

[33]  Álisson Rocha Machado,et al.  Application of cutting fluids in machining processes , 2001 .

[34]  Andrea Ko,et al.  Monitoring Pneumatic Actuators’ Behavior Using Real-World Data Set , 2020, SN Computer Science.

[35]  Álisson Rocha Machado,et al.  Performance of cryogenically treated HSS tools , 2006 .

[36]  Andrés Bustillo,et al.  Identifying maximum imbalance in datasets for fault diagnosis of gearboxes , 2018, J. Intell. Manuf..

[37]  Meng Ma,et al.  Extract interpretability-accuracy balanced rules from artificial neural networks: A review , 2020, Neurocomputing.

[38]  Ulvi Şeker,et al.  Experimental investigation of the effect of cutting tool rake angle on main cutting force , 2005 .

[39]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[40]  Guillem Quintana,et al.  Avoiding neural network fine tuning by using ensemble learning: application to ball-end milling operations , 2011 .

[41]  Andres Bustillo,et al.  The evolutionary development of roughness prediction models , 2013, Appl. Soft Comput..

[42]  Andrés Bustillo,et al.  Semi-supervised roughness prediction with partly unlabeled vibration data streams , 2018, Journal of Intelligent Manufacturing.

[43]  Franci Cus,et al.  Optimization of cutting conditions during cutting by using neural networks , 2003 .

[44]  J. Antony,et al.  Analysis and Interpretation of Data from Taguchi Experiments , 2000 .

[45]  Ponnuthurai N. Suganthan,et al.  Ensemble Classification and Regression-Recent Developments, Applications and Future Directions [Review Article] , 2016, IEEE Computational Intelligence Magazine.

[46]  L. N. López de Lacalle,et al.  Effects of high-pressure cooling on the wear patterns on turning inserts used on alloy IN718 , 2017 .

[47]  Lutfi Taner Tunc,et al.  Effect of cutting conditions and tool geometry on process damping in machining , 2012 .

[48]  Joaquim Ciurana,et al.  Modeling pulsed laser micromachining of micro geometries using machine-learning techniques , 2015, J. Intell. Manuf..

[49]  Kai Cheng,et al.  Investigation of influence of tool rake angle in single point diamond turning of silicon , 2018 .

[50]  Adam G. Rehorn,et al.  State-of-the-art methods and results in tool condition monitoring: a review , 2005 .

[51]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[52]  Nouredine Ouelaa,et al.  Experimental study of the combined influence of the tool geometry parameters on the cutting forces and tool vibrations , 2015 .

[53]  Andrés Bustillo,et al.  Artificial intelligence for automatic prediction of required surface roughness by monitoring wear on face mill teeth , 2017, Journal of Intelligent Manufacturing.

[54]  Taylan Altan,et al.  Effect of tool edge geometry and cutting conditions on experimental and simulated chip morphology in orthogonal hard turning of 100Cr6 steel , 2009 .

[55]  Moola Mohan Reddy,et al.  Environmental friendly cutting fluids and cooling techniques in machining: a review , 2014 .

[56]  Uday S. Dixit,et al.  Application of soft computing techniques in machining performance prediction and optimization: a literature review , 2010 .

[57]  Á. R. Machado,et al.  2.9 Heat Treatment of Tool Steels , 2017 .

[58]  Joaquim Ciurana,et al.  Random Forest ensemble prediction of stent dimensions in microfabrication processes , 2017 .