Machine-learning for automatic prediction of flatness deviation considering the wear of the face mill teeth
- PDF / 2,056,587 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 27 Downloads / 179 Views
Machine-learning for automatic prediction of flatness deviation considering the wear of the face mill teeth Andres Bustillo1
· Danil Yu. Pimenov2
· Mozammel Mia3
· Wojciech Kapłonek4
Received: 26 February 2019 / Accepted: 13 August 2020 © The Author(s) 2020
Abstract The acceptance of the machined surfaces not only depends on roughness parameters but also in the flatness deviation (fl ). Hence, before reaching the threshold of flatness deviation caused by the wear of the face mill, the tool inserts need to be changed to avoid the expected product rejection. As current CNC machines have the facility to track, in real-time, the main drive power, the present study utilizes this facility to predict the flatness deviation—with proper consideration to the amount of wear of cutting tool insert’s edge. The prediction of deviation from flatness is evaluated as a regression and a classification problem, while different machine-learning techniques like Multilayer Perceptrons, Radial Basis Functions Networks, Decision Trees and Random Forest ensembles have been examined. Finally, Random Forest ensembles combined with Synthetic Minority Over-sampling Technique (SMOTE) balancing technique showed the highest performance when the flatness levels are discretized taking into account industrial requirements. The SMOTE balancing technique resulted in a very useful strategy to avoid the strong limitations that small experiment datasets produce in the accuracy of machine-learning models. Keywords Face milling · Wear · Tool life · Tool condition monitoring · Flatness deviation · Cutting power · Random forest · SMOTE
List of symbols CNC Computer numerical control SMOTE Synthetic Minority Over-sampling Technique
B B
Danil Yu. Pimenov [email protected] Wojciech Kapłonek [email protected] Andres Bustillo [email protected] Mozammel Mia [email protected]
1
Department of Civil Engineering, Universidad de Burgos, Avda Cantabria s/n, 09006 Burgos, Spain
2
Department of Automated Mechanical Engineering, South Ural State University, Lenin Prosp. 76, Chelyabinsk, Russia 454080
3
Department of Mechanical Engineering, Imperial College London, London SW7 2AZ, UK
4
Department of Production Engineering, Faculty of Mechanical Engineering, Koszalin University of Technology, Racławicka 15-16, 75-620 Koszalin, Poland
t T ap B fz vc n VB Pc fl χYDeg L B H D γ α λ kr k r1
Tool life (processing time) (min) Work cycle time (machining pass) (min) Cutting depth (mm) Cutting width (mm) Feed per tooth (feed rate) (mm/tooth) Cutting speed (m/min) Spindle rotation speed per minute (rpm) Tool flank wear value on the flank surface (flank wear) (mm) Drive power primary motion (kW) Flatness deviation (μm) Torsional angle of a machine tool-device-spindle (°) Workpiece length (mm) Workpiece width (mm) Workpiece height (mm) Cutter diameter (mm) Rake angle (°) Back angle (°) Angle of the main cutting edge (°) Major cutting edge angle (°) Minor cutting edge angle (°)
123
Journal of Intelligent Manufacturing
z k
Number of teeth Number
Data Loading...
