Intelligent PCD Tool Testing and Prediction of Performance

Abstract

The ongoing competition on cutting tools drive the tool manufactures to enhance the cutting tool quality through real-time condition monitoring and performance prediction. The digitization of tool within the manufacturing process, allows for the creation of an intelligent manufacturing system, to reduce the cycle time required for testing, and facilitate the categorization of tool performance as well as early warning signals of the change in the manufacturing processes. The system comprises two platforms such as feature extraction engine (FEE) and feature prediction engine (FPE). The FEE monitors real-time progression of operational behavior during wear with the tool–work interface signals. The accelerated tool wear testing applies a milling arrangement, incorporating a clock-testing workpiece that simulates an intermittent cutting process on hardened steel. The feature extraction engine uses the accelerated wear results to build a calibrated wear model as a reference tool for wear analysis and prediction. Flank wear lands were imaged using a Leica toolmaker’s microscope and used to calibrate the wear model, correlating the digital signal feature to the tool–work interface wear behavior. The imaged wear progression, force, and acoustic emission signal features were analyzed by statistical methods including applications of Spiro-Wilks, ANOVA, and Kruskal–Wallis evaluations. This confirms the experimental accuracy and provides the baseline for wear prediction driving the development of a feature prediction engine (FPE). The results are a significant reduction in the quality control cycle time and performance prediction. The experimental results indicate that the FEE correlates accurately across sensors and progression tracking of abrasive wear in the cutting tools, clearly distinguishing between machining cuts, signal noise, and signal anomalies.

Appendix 1: Computation of Volumetric Wear Using the Principles Cylindrical Wedge

A wedge is cut from a cylinder by slicing with a plane that intersects the base of the cylinder. The volume of a cylindrical wedge can be found by noting that the plane cutting the cylinder passes through the three points illustrated above (with b > R), so the three-point form of the plane gives the equation

$$ \left| {\begin{array}{*{20}c} x & y & z & 1 \\ {R - b} & a & 0 & 1 \\ {R - b} & { - a} & 0 & 1 \\ R & 0 & h & 1 \\ \end{array} } \right| = h(R - b - x) + bz $$

Solving for z is given as

$$ z = \frac{(h)(x - R + b)}{b} $$

The value of “a” is given as

$$ \begin{aligned} a & = \sqrt {\left( {R^{2} } \right)\left( {(b - R)^{2} } \right)} \\ a & = \sqrt {b(2R - b)} \\ \end{aligned} $$

The volume of cylindrical wedge is given as the integral of rectangular areas along the x-axis

$$ \begin{aligned} z & = \int {Z(x)y(x)} \,{\text{d}}x \\ Z & = \int_{R - b}^{R} {\frac{h(x - R + b)}{b}\sqrt {\left( {R^{2} - x^{2} } \right)} } \,{\text{d}}x \\ z & = \frac{h}{6b}\left\lfloor {2\sqrt {(2R - b)b } \left( {3R^{2} - 2ab + b^{2} } \right) - 3\pi R^{2} (R - b)\sin^{ - 1} \frac{(R - b)}{R}} \right\rfloor \\ a & = R\sin \theta ;\quad b = R(1 - \cos \theta );\;\;b^{2} = 2bR - a^{2} \\ \end{aligned} $$

Hence, $V = \frac{2}{3}hR^{2}$.