Abstract:In the mineral processing process, the detection of ore lump size distribution is usually realized by manual sampling and screening. There exist issues like large labor consumption and inability to detect the ore lump size of the mill online in real time. Besides, there is also need to improve the grinding index, save energy and reduce consumption, and finally achieve the purpose of optimal control. In this paper, high-speed image acquisition technology is applied to monitor ore lumps at the feeding opening of the conveyor belt in real time, and the high-definition ore images are segmented via image segmentation technology. The F80 value is transmitted to the intelligent control system in real time, and the development and application of the real-time lumpiness analysis system for the feed of the mill is completed. The practical application shows that the results of ore particle size analysis can achieve granular control over the grinding process of semi-autogenous grinding, maximize the processing capacity of the mill, ensure the particle size of the grinding products, improve the grinding performance and indicators, and reduce the labor intensity of operators.