In this work, we propose a capacitive load cell for detecting changes in material and tool wear conditions for conical picks used in underground mining. Currently, material type and tool wear are primarily deduced by human operators near the cutting interface. This process is dangerous and subjective, which makes training difficult. Sensors that can detect material and tool wear conditions stand to improve worker safety by enabling operators to perform their functions from a greater distance. We have developed a capacitive load cell that integrates with the target tooling and can detect changes in vibrational modes which correlate with the different material and wear conditions. To test our device, we installed it onto a conical pick in the linear cutting machine at the Earth Mechanics Institute of the Colorado School of Mines. Three different wear levels, ranging from new to worn, were tested with a limestone block embedded in concrete. Measurements from the sensor and from strain gauges embedded in the linear cutting machine were collected. To classify the samples based on modal differences, a support-vector machine was used with the magnitude of the Fourier spectra of windows of data from the different conditions. This method reliably classifies samples under our laboratory conditions, indicating it is suitable for consideration for further integration with cutting implements.
Full-text paper:
Mining, Metallurgy & Exploration (2023) 40:757–771, https://doi.org/10.1007/s42461-023-00732-2
