Research On Analysis Method Of AOD Furnace Lining Erosion

AOD furnace is the key equipment for ferroalloy smelting.In the smelting process,the furnace temperature can reach 1700°C,or even higher.With complicated chemical reactions,the lining erosion is more serious and it is one of the important factors affecting the service life of AOD furnace.The timely and accurate measurement and analysis of the lining erosion can improve the lining repair program,improve the overall service life of the lining,and prevent production accidents.At present,compared with common AOD furnace detection technologies such as laser detection,ultrasonic detection,and infrared detection at home and abroad,it is rare to see the application of three-dimensional visual inspection methods to detect the extent of lining erosion.This article adopts three-dimensional visual inspection technology and realizes automatic measurement and analysis of the lining erosion of AOD furnace based on the three-dimensional CAD software platform.It lays an important theoretical foundation for further improving the production process,improving the furnace service,and highlighting the advantages of AOD furnace in energy saving and consumption reduction.This article uses the Kinect2.0 sensor as an image data acquisition tool.The filtering,streamlining,and surface reconstruction and restoration of the acquired image data are processed,and the amount of lining erosion is measured by the secondary development of the three-dimensional CAD software.The main research contents of this paper are as follows:(1)Combine the structure and parameters of AOD furnace,analyze the lining erosion through AOD furnace smelting mechanism,study the characteristics of the erosion zone of tuyere and slag line in AOD furnace;(2)Give a removal Outliers and moving least squares joint filtering method.Combined with the principle of outlier removal based on the average distance and number of neighborhoods,the filtering effect of each method on the research object was verified through experiments.On this basis,an outlier removal method of image data was given and determined related parameters through experiments.The data was smoothed by moving least-squares method.The relative parameters of the smoothing algorithm were given by comparing the experiment before and after the processing.(3)Analyze the characteristics of the data reduction method based on the normal vector and the curvature,adopt the voxel grid to divide the data,calculate the center of gravity of the data points in the voxel grid,and use the center of gravity to replace other data points in the grid.The data was streamlined and the length of the voxel grid was determined experimentally,and the effectiveness of the method was verified.(4)The principles of the Poisson surface reconstruction and greedy projection surface reconstruction algorithm were analyzed in depth.After comparing and analyzing the reconstruction effects of the algorithm,The Poisson surface reconstruction algorithm is used to determine the recursive depth parameters that have a significant influence on the reconstruction time of the surface.Experiments show that the algorithm has a good reconstruction effect and is time-consuming;(5)Using the VS development platform,The CATIA V5 software and CAA components realized the secondary development of CATIA,combined with the surface reconstruction model and the external wall parameter model of the AOD furnace to construct a lining erosion measurement and analysis platform.