A Novel Computer Vision Based Machine Learning Approach For Online Tool Wear Monitoring In Machining

With the increased scope of automated machining processes, one of the essential requirements is the reliable predictions of the tool life. It is crucial to monitor the condition of the cutting tool during the machining process to achieve high-quality machining and cost-effective production. This paper presents a computer vision technique for flank wear measurement and prediction using machine learning, specifically support vector machine (SVM) and boosted decision trees has been used. The proposed methodology for tool wear measurement is illustrated for the CNC machining experimentally. The direct method of tool wear measurement and prediction have been proposed. Flank wear measurement is carried out on PVD coated tool insert, and experiments are performed on an alloy steel workpiece under dry machining. For capturing images, a CMOS camera with a lens mounted on the machine is used. To avoid environmental effects on the images LED ring light is used. Captured tool insert images are provided to the image processing algorithm built-in MATLAB software. The measurement of flank wear is also carried out using a microscope. The prediction accuracy of SVM and optimized boosted tree models is 97% and 96%, respectively, proving prediction algorithms' effectiveness. The findings showcased that the proposed methodology can measure and predict the tool wear with higher accuracy. It has demonstrated the ability to increase cutting tool utilization with the improved surface finish of the machined component.

Measurement of Tooth Wear by Means of Digital Impressions: An In-Vitro Evaluation of Three Intraoral Scanning Systems

This in-vitro study aimed to investigate whether intraoral scanners (IOS) are suitable for wear measurement compared to optical profilometry (WLP). A zirconia cast representing the teeth (24–28) was fabricated. It was digitized six times using three different intraoral scanners, Cerec Omnicam AC (OC), Trios 3 (Tr3), and True Definition (TD). The scans were conducted at baseline (t0) and at three different stages of simulated wear (t1–t3), each at one wear-facet on FDI 26 and FDI 27. WLP was used as a reference method. Within each acquisition system, the maximum wear at each facet was analyzed by superimposing the STL data of t0 with t1–t3. A power analysis was performed (G*Power), and the Wilcoxon-signed-rank-test was used to evaluate whether there were statistically significant differences between the groups (Bonferroni corrected) (α = 0.05). At wear-facet FDI 27, differences from +4% t1 TD up to +19% t2 OC, corresponding to a metric value of 8 µm and 45 µm, were measured. At FDI 26 deviations between −2% t1 Tr3, and +10% OC and Tr3, were observed. Considering some limitations, the IOS are a promising alternative to wear measurement based on WLP due to its simple application to capture surface changes in a reasonable and quick way.

Industrial Vertical Stirred Mills Screw Liner Wear Profile Compared to Discrete Element Method Simulations

Vertical stirred mills have been widely applied in the minerals industry, due to its greater efficiency in comparison with conventional tumbling mills. In this context, the agitator liner wear plays an important role in maintenance planning and operational costs. In this paper, we use the discrete element method (DEM) wear simulation to evaluate the screw liner wear. Three different mill rotational velocities are evaluated in the simulation, according to different scale-up procedures. The wear profile, wear measurement, power consumption, and particle contact information are used for obtaining a better understanding of the wear behavior and its effects on grinding mechanisms. Data from a vertical stirred mill screw liner wear measurement obtained in a full-scale mill are used to correlate with simulation results. The results indicate a relative agreement with industrial measurement in most of the liner lifecycle, when using a proper mill velocity scale-up.