HJKIR HomePage
The tail of the sintering machine is the end point of the sintering process. The degree of burn through of the sintering deposit at the tail of the machine directly reflects whether the running speed of the sintering machine is reasonable, the permeability of the material layer and the fuel ratio. It is one of the important qualitative indicators for production regulation.
For a long time, in order to realize the continuous stability and high-quality production of sintering production, the methods of direct observation manually and observation through the visible light industrial television system at the end of the sintering machine have been mainly adopted. The method of analyzing the infrared image of the section at the end of the sintering machine is generally used for the online monitoring of sinter quality.
The infrared image of the sinter cake section in the discharge area at the end of the sintering machine can directly reflect the information of the material layer state. It is a comprehensive reaction of the sintering production process and can be used as the main basis for controlling the heat level in the sintering material layer. In the past, the operators used the tail section to observe the brightness, color, size, distribution and position of the red ribbon in the adjacent tail sections of several sintering machines, the integrity of sinter cake unloading and the size of dust, etc., and combined their own experience and knowledge to judge the quality index of sintering and the reasonable value of various operating parameters. However, due to the interference and incompleteness of human factors in this method, people seek to develop some simulation operators' work to varying degrees, and use the continuous acquisition and judgment system of the image of the tail section of the sintering machine to overcome the shortcomings of the artificial method.
The prediction model of sintering end point is to accurately form the red layer image of the end section of the sintering machine by means of thermal imaging. Image preprocessing is used to extract the characteristics of the red flame layer in the end section image of the sintering machine, such as the area, center of gravity, perimeter and so on, and extract the characteristic parameters such as the average brightness and average area value of the sintering pores related to FeO content. By establishing the inference rules related to the characteristics of the end section image of the sintering machine and combining the extracted characteristic parameters, By comparing the brightness of the red flame layer between the frames of the sintering machine tail video, the key frames of the sintering machine tail section can be found accurately. Display the sintering quality information and image at the same time. At the same time, the decision-making information of sintering control production is given by using data analysis, FeO prediction and other technologies to guide the production control operation of sintering end-point model and ignition optimization model.
The specific functions are as follows:
1) Infrared thermal image processing display
The image effect suitable for the environment of the sintering machine tail is adopted to realize the display function of continuous images, and create good conditions for the subsequent feature image acquisition, the analysis of the temperature field of the sinter deposit section and the online inference of the sinter quality index without distortion.
2) Feature image acquisition
During the continuous operation of the sintering machine, the section state of each lathe layer is screened to capture a clear picture, a complete section of the sintering deposit, and a representative section image of the tail of the machine that can better describe the distribution of the thermal state in the bed, so as to ensure the effectiveness and accuracy of the analysis of the thermal state of sintering. Section feature image screen display.
3) Online inference of variation trend of sinter quality index
Extract the image features of sinter quality indicators in the sinter deposit section feature image, and adopt effective and practical data compression and feature optimization methods for data processing. On this basis, the online inference model of FeO change trend of sinter is established for online identification and analysis.
4) The FeO content of sinter is preliminarily judged through the analysis of the section at the tail of the machine.
