Research On The Contour Identification And Intelligent Optimization For Cutting Strategy Of Wide-Heavy Plate

With the continuous advancement of science and technology,intelligent manufacturing has become the only way for the development of the manufacturing industry.This is the main direction for the implementation of "Made in China 2025" and also the core content of China's steel industry to enhance its competitiveness.In this regard,this topic is aimed at the problem of inaccurate division of shear lines in a wide-heavy plate production line of a steel company in China.In order to improve the yield of wide-heavy plates,a set of wide-heavy plate contour recognition and intelligent shearing system based on machine vision was developed.The system can quickly and accurately identify the characteristics of the tail and side bends of the wide-heavy plates,and combines intelligent algorithms to optimize the calculation of the cutting line to reduce the loss of the wide-heavy plate in shearing process.The main research contents and results of the thesis are as follows:(1)Based on the characteristics of the wide-heavy plate's cutting line,the structure of the plate contour detection system was completed.After the equipment selection of the contour detection system,a simulated test bench was set up in the laboratory;By introducing the distortion model into the camera imaging model,the calibration of the camera and the distortion correction of the image are completed.(2)A set of preprocessing flow based on image gray stretch,image median filtering,Otsu's threshold segmentation and generalized image morphology operation is designed;By comparing the related image contour extraction algorithm,a method combining the Canny operator and the grayscale gradient interpolation sub-pixel edge contour extraction algorithm is proposed.In the case of CPU is i5,the frequency is 2.30GHz,the RAM is 8GB,and the image resolution is 2056×2464,The whole image processing algorithm runs for less than 1s,and the detection accuracy can reach 0.2 pixels.(3)Established a wide-heavy plate shear line division model.Firstly,using affine transformation to tilt the image contour data to improve the accuracy of the measurement;Then,based on the scan line strategy,the feature recognition of the steel plate contour image is proposed,and different shear strategies are set according to different identification results;Finally,a hybrid genetic layout algorithm is proposed for the limitation of the wide-heavy plate shearing process.The field data is used for testing and compared with the original cutting method.The results show that the utilization rate of the steel plate using the optimized layout strategy is 10%higher than the original cutting method.(4)The design of the contour extraction and cutting system software for the plate was completed,and the sample was tested with system software.The experimental test results show that the measurement error of the characteristics of the wide-heavy plate samples are not more than 0.5mm.