Sintering Based On The Image Processing Section Temperature Measurement

This paper is supported by the Natural Science Foundation of He'nan Province and Opening Foundation of National Laboratory of Pattern Recognition , Chinese Academy of Sciences.Sintering is an important segment that provides raw materials for the blast furnace. Good quality , high production and energy consumption in iron-making are directly influenced by the high or low production and the quality of sintering ores. Sintering is a complicated dynamic process with more than ten kinds of physic-chemistry changes, twenty state variables and nearly eighty related factors. At present, sintering terminal is controlled indirectly by people observing conventional instrument and industrial television. But this controlling means has some shortcomings such as bad productive stability, more return-ore quantity and short cycle of maintenance of equipment etc. Some scholars who study sintering technology show that it couldn't resolve the puzzle only through the general statistic model or mechanism model, and it must depend on information science technology.The research shows that we can get more direct and abundant information about production of sintering from the cross-section image of the discharge end. According to sintering technology, the sintering process includes hot states all the time. Temperature is an important parameter in many parameters of sintering. It is related to return-ore balance, reducibility, mechanical robustness and utilized coefficient. So temperature field measurement of the cross-section plays an important role in realizing automatic control of sintering process.Around these problems such as optimum control of sintering and temperature field measurement of the cross-section, this paper presented the scheme of the intelligent self-adapting control system of sintering and provided software diagram and hardware allocation through using pattern recognition, neural network and self-adapting control technology etc; during the normal running conditions, this paper presented a new method of non-contact temperature measurement by digital image processing and infrared temperature measurement technology. It will provide a new method of sinteringterminal control.The main work is as follows:1. The design of the intelligent self-adapting control system of sintering, such as having analyzed and compared several important methods of sintering production; putting forward the important research value of the sintering production; providing the design of the whole system, the flow process diagram and hardware components etc.2. Pre-processing of the cross-section of the discharge end. Based on the exactly gathering a frame of image, studying three kinds of image correction methods, using bivariate cubic polynomial effectually corrects the image; using adaptive neighborhood mean method smoothes the image by contrast; studying two operators and wavelet theory detect the margin of the cross-section image, but using the wavelet theory shows the good effect.3. The method of calibration and measurement of temperature field of the cross-section of the discharge end. Researching the method and the theory of temperature measurement through using images, it is mainly based on the corresponding relation of the thermal radiation and the gray level, this paper is adopted the total radiation method chiefly based on the Planck radiation law. What's more, it deduces the formula of temperature measurement by the total radiation law and amends the formula according to practical situation, then shows the method of acquiring the temperature field of the cross-section images of sintering by using regression analysis.4. Having given some measurements to improve the precision of temperature detecting, building the foundation for the field engineering application.