Technical vision system for quartz raw material quality assessment

One of the key problems of metallurgical production is the quality fluctuations of feedstock such as ores, concentrates and metallurgical charge. The chemical and mineralogical composition of the ore and their stability in the ore flow remain the main ones when assessing the significance of indicators characterizing the quality of the mineral in general. Analysis of production data of industrial silicon smelting in ore thermal furnaces has shown that the efficiency of the technological process, the volume of dust emissions and the purity of the final product depend on the quality of quartz raw materials. Ore quality is determined by such functional characteristics as percentage of harmful impurities, strength of ore mass, mineral and structural composition, moisture, etc. Stability of ore mass composition has the most significant influence on the results of metallurgical production. To solve the problem of accounting for fluctuations in the chemical composition of quartzite, the use of technical vision system is relevant. The paper describes an automatic system for analyzing the quality of quartz raw material for the production of metallurgical silicon based on the evaluation of visual characteristics of quartzite. The developed algorithm is based on colour filtering of images of mineral cuts. The result of the work is the data on the mass amount of iron impurity inclusions, and the average reliability of the estimation correlates with the data of X-ray spectral analysis based on the test results on average by 90%. Positive effects of implementation of the proposed vision system are the efficiency of quality assessment during unloading and preparation of raw materials, which allows taking measures to adjust the proportional composition of the charge of the ore thermal furnace.