Machine learning based approach for process supervision to predict tool wear during machining

Numerische Steuerungen für Werkzeugmaschinen erfassen eine erhebliche Menge an Sensordaten für die Achsregelung. Diese liefern nicht nur Informationen über die aktuellen Achspositionen oder die Ströme, sondern können mithilfe von Modellen auch für das Monitoring von anderen Prozessgrößen verwendet werden. In diesem Beitrag wird ein Machine-Learning-Verfahren zur Überwachung von Werkzeugverschleiß untersucht, welches allein auf maschinen-internen Daten basiert.   Numerical controls for machine tools acquire a considerable amount of sensor data for axis control. This information, such as the current axis position or the motor currents, can be used for monitoring other process variables with the aid of models. This article investigates a machine learning method for monitoring tool wear in machine tools, based on machine-internal data only.

Download Full-text

Machine Learning Data Analysis for Tool Wear Prediction in Core Multi Process Machining

Journal of the Korean Society of Manufacturing Process Engineers ◽

10.14775/ksmpe.2021.20.09.090 ◽

2021 ◽

Vol 20 (9) ◽

pp. 90-96

Keyword(s):

Machine Learning ◽

Data Analysis ◽

Tool Wear ◽

Wear Prediction ◽

Tool Wear Prediction ◽

Learning Data

Download Full-text

Predicting the Tool Wear of a Drilling Process Using Novel Machine Learning XGBoost-SDA

Materials ◽

10.3390/ma13214952 ◽

2020 ◽

Vol 13 (21) ◽

pp. 4952

Author(s):

Mahdi S. Alajmi ◽

Abdullah M. Almeshal

Keyword(s):

Machine Learning ◽

Cast Iron ◽

Tool Wear ◽

Flank Wear ◽

Accurate Prediction ◽

Superior Performance ◽

Gradient Boosting ◽

Support Vector ◽

Drilling Process ◽

Extreme Gradient Boosting

Tool wear negatively impacts the quality of workpieces produced by the drilling process. Accurate prediction of tool wear enables the operator to maintain the machine at the required level of performance. This research presents a novel hybrid machine learning approach for predicting the tool wear in a drilling process. The proposed approach is based on optimizing the extreme gradient boosting algorithm’s hyperparameters by a spiral dynamic optimization algorithm (XGBoost-SDA). Simulations were carried out on copper and cast-iron datasets with a high degree of accuracy. Further comparative analyses were performed with support vector machines (SVM) and multilayer perceptron artificial neural networks (MLP-ANN), where XGBoost-SDA showed superior performance with regard to the method. Simulations revealed that XGBoost-SDA results in the accurate prediction of flank wear in the drilling process with mean absolute error (MAE) = 4.67%, MAE = 5.32%, and coefficient of determination R2 = 0.9973 for the copper workpiece. Similarly, for the cast iron workpiece, XGBoost-SDA resulted in surface roughness predictions with MAE = 5.25%, root mean square error (RMSE) = 6.49%, and R2 = 0.975, which closely agree with the measured values. Performance comparisons between SVM, MLP-ANN, and XGBoost-SDA show that XGBoost-SDA is an effective method that can ensure high predictive accuracy about flank wear values in a drilling process.

Download Full-text

Tool wear estimation in turning of Inconel 718 based on wavelet sensor signal analysis and machine learning paradigms

Production Engineering ◽

10.1007/s11740-020-00989-2 ◽

2020 ◽

Vol 14 (5-6) ◽

pp. 693-705

Author(s):

Tiziana Segreto ◽

Doriana D’Addona ◽

Roberto Teti

Keyword(s):

Machine Learning ◽

Tool Wear ◽

Inconel 718 ◽

Signal Analysis ◽

Elevated Temperatures ◽

Wavelet Packet ◽

Research Work ◽

High Strength ◽

Signal Features ◽

Tool Wear Estimation

AbstractIn the last years, hard-to-machine nickel-based alloys have been widely employed in the aerospace industry for their properties of high strength, excellent resistance to corrosion and oxidation, and long creep life at elevated temperatures. As the machinability of these materials is quite low due to high cutting forces, high temperature development and strong work hardening, during machining the cutting tool conditions tend to rapidly deteriorate. Thus, tool health monitoring systems are highly desired to improve tool life and increase productivity. This research work focuses on tool wear estimation during turning of Inconel 718 using wavelet packet transform (WPT) signal analysis and machine learning paradigms. A multiple sensor monitoring system, based on the detection of cutting force, acoustic emission and vibration acceleration signals, was employed during experimental turning trials. The detected sensor signals were subjected to WPT decomposition to extract diverse signal features. The most relevant features were then selected, using correlation measurements, in order to be utilized in artificial neural network based machine learning paradigms for tool wear estimation.

Download Full-text

A computer vision approach to analyze and classify tool wear level in milling processes using shape descriptors and machine learning techniques

The International Journal of Advanced Manufacturing Technology ◽

10.1007/s00170-016-9541-0 ◽

2016 ◽

Vol 90 (5-8) ◽

pp. 1947-1961 ◽

Cited By ~ 16

Author(s):

María Teresa García-Ordás ◽

Enrique Alegre ◽

Víctor González-Castro ◽

Rocío Alaiz-Rodríguez

Keyword(s):

Machine Learning ◽

Computer Vision ◽

Tool Wear ◽

Machine Learning Techniques ◽

Shape Descriptors ◽

Learning Techniques

Download Full-text

Application of Audible Sound Signals for Tool Wear Monitoring and Workpiece Hardness Identification in Gear Milling Using Machine Learning Techniques

Volume 10: 2017 ASME International Power Transmission and Gearing Conference ◽

10.1115/detc2017-68067 ◽

2017 ◽

Author(s):

Achyuth Kothuru ◽

Sai Prasad Nooka ◽

Patricia Iglesias Victoria ◽

Rui Liu

Keyword(s):

Machine Learning ◽

Tool Wear ◽

Learning Algorithm ◽

Machining Process ◽

Machine Learning Techniques ◽

Support Vector ◽

Process Conditions ◽

Tool Wear Monitoring ◽

Audible Sound ◽

Wear Monitoring

The machining process monitoring, especially the tool wear monitoring, is very critical in modern automated gear machining environment which needs instant detection of cutting tool state and/or process conditions, quick final diagnosis and appropriate actions. It has been realized that the non-uniform hardness of the workpiece material due to the improper heat treatment can cause expedited tool wear and unexpected tool breakage, which greatly increases difficulties and complexities in monitoring the tool conditions in gear cutting. This paper provides a solution to detect the wear conditions of the gear milling cutter in the cutting of workpiece materials with hardness variations using the audible sound signals. In this study, cutting tools and workpieces are prepared to have different flank wear classes and hardness variations respectively. A series of gear milling experiments are operated with a broad range of cutting conditions to collect sound signals. A machine learning algorithm that incorporates support vector machine (SVM) approach coupled with the application of time and frequency domain analysis is developed to correlate observed sound signals’ signatures to specified tool wear classes and workpiece hardness levels. The performance evaluation results of the proposed monitoring system have shown accurate predictions in detecting tool wear conditions and workpiece hardness variations from the sound signals in gear milling.

Download Full-text