Inductive Sensing for the Differentiation of Superior and Inferior Copper Ores

This article presents a novel inductive sensing system that is integrated in an excavation bucket to assess the quality of copper ore at the mining site and whether it justifies transportation and further processing. The mining industry lacks low-cost and reliable nonintrusive sensing assessment systems for this purpose. Given that the quality of copper ore is a function of its conductivity, we propose using an inductive sensing approach to develop a new sensor. We designed suitable sensing coils and measured the real and imaginary parts of the coils at a specific excitation frequency when ores of different qualities were in their direct vicinity. From these measurements, superior and inferior quality ores were differentiated. The standard deviation of the measurements conducted is 0.0112. To differentiate superior from inferior ores, a threshold is defined by adding three times the standard deviation to the mean of repeated readings for superior ore samples. Assuming a normal distribution, this approach yields a classification accuracy of approximately 99.73%. For even greater precision, the designer may opt to raise the threshold further. The proposed sensor, integrated into an excavation bucket, was tested in a laboratory environment for level sensing and distinguishing good ore from bad ore. Moisture testing showed that the sensor is negligibly affected. The sensing system uses three sets of coils. The middle planar coil (MPC) is a series combination of multiple coils to sense the fill level to determine the volume of ore collected and check whether sufficient ore is available for quality measurement. The right planar coil (RPC) and left planar coil (LPC) sensors are introduced at the bottom of the excavation bucket to sense the ore quality information. Enabling on-site grading of copper ore enhances economic feasibility, minimizes further waste generation, and reduces resource consumption. The sensing approach and the system are evaluated using finite element analysis (FEA) and experimental studies on a developed prototype. The results strongly support the usefulness of the proposed sensing technique.