Research On 3D Reconstruction And Sag Detection Of Low-Temperature Adiabatic Cylinder Appearance

Cryogenic liquid has the characteristics of easy gasification and ultra-low temperature,etc.,and it is particularly important to store it safely when it is widely used in industrial production and daily life.In recent years,the storage of cryogenic liquid mainly uses low-temperature adiabatic cylinder.The structure of low-temperature adiabatic cylinder is characterized by vacuum between the inner and outer walls,which can improve the adiabatic performance of the cylinder under the premise of ensuring the vacuum degree.When defects appear on the surface of low-temperature adiabatic cylinder,the adiabatic performance of the cylinder will be affected.In serious cases,large amount of cryogenic liquid gasification in gas cylinders will appear overpressure phenomenon,which will easily cause explosion,frostbite and other accidents.At present,the defect detection on the surface of low-temperature adiabatic cylinders is mainly by manual,but the efficiency is low and the probability is large.Therefore,in order to ensure the safe operation of low-temperature adiabatic gas cylinders,relying on the "Three-dimensional Reconstruction and Sag Detection Project of the Low-temperature Adiabatic Cylinder Appearance in Cryogenic Insulation Equipment Product Quality Supervision Inspection of X Province",carried out the three-dimensional reconstruction of the appearance of low-temperature adiabatic cylinders and the research of sag identification.Firstly,the detection accuracy of the system is affected by the baseline distance,magnification and the angle between the laser plane and the CCD camera.At the same time,the three-dimensional reconstruction and dent detection system structure of low-temperature adiabatic cylinder appearance with the best detection accuracy is built based on the field situation.Secondly,3D detection method based on linear structured light is used to identify and detect the depression of cylinder surface.The linear structured light is transmitted to the surface of the low-temperature adiabatic cylinder and the image information is acquired by the vision sensor.Because the distribution of the structural light bar in the image is relatively fixed,the image is segmented to improve the detection efficiency,and the center point of the structural light bar is extracted by using the gray distribution of row direction.Lastly,as a constraint condition of 3d reduction process,the calibration of structured light plane further improves the accuracy of transformation from 2d coordinate information to 3d point cloud data.Driven by the servo motor motion control system,the vision sensor completed the overall scanning of the low-temperature adiabatic gas cylinder,and the calibration of the scanning direction improved the accuracy of the single point cloud data set.Three sets of point cloud data sets were spliced,denoised,smoothed,cavity patched and rendered to obtain the overall three-dimensional model of the low-temperature adiabatic cylinder.Based on the basic sag detection,this paper proposes an optimized detection algorithm,which can realize the detection of sag position,depth,circumferential width and axial length.Compared with the basic detection method,the detection accuracy is greatly improved,which has certain engineering significance.