| Iron ore pellets are one of the main materials in the steel industry.Iron ore pellets are prepared from green pellets after drying,firing and cooling processes.At present,disc pelletizers are often used in industry to prepare green pellets with a certain moisture content.The moisture content of green pellets is an important quality index in the pelletizing process.However,at present,the moisture content of green pellets are mostly measured offline by manual,which is low efficient and time-lag.Therefore,there is an urgent need to develop on-line measuring method for the moisture content of green pellets to improve the automation level of the pelletizing process and the product quality.Aimed at the above problems and funded by National Natural Science Foundation of China,an image-based method is studied in the present work for online measurement of the moisture content of green pellets.This work helps to improve the automatic control of the pelletizing process,and has high theoretical significance and engineering value.The main contributions of this paper are as follows:(1)An experimental system was established to capture images of green pellets.A large amount of images under different experimental configurations(intensity of light,the placement of the green pellets and the amount of water spraying)were collected and a data set was established;(2)The images are processed to extract features related to moisture content in green pellets.Then a regression model between the image skewness and the moisture content was established,with which the measurement of moisture content can be accomplished.(3)The proposed method was verified by experiments.The results show that,compared with the grayscale based method,the proposed method has higher accuracy in measuring moisture content in green pellets,with measuring error can meet the actual requirements of industrial site.Moreover,it has better robustness to the changes of image illumination and the placement of the green pellets;(4)The proposed measuring method is verified by using green pellets image collected from pelletizing process in a local steel plant.The results show that it correctly reflects the trend of moisture content changes under different process conditions.Moreover,the computing speed is fast enough to meet the requirement for real-time applications.(5)The proposed method was implemented in the platform Visual Studio2013 +Qt5.8 and C ++.A software was developed for on-line measurement of moisture content in green pellets.The proposed method is useful in on-line measurement of the moisture content of green pellets,and helps to improve the automatization level of the pelleting process and the product quality. |
PDF Full Text Request