Hot-rolled strip steel optimized shearing method based on machine vision

Abstract

The invention belongs to the field of hot-rolled strip steel flying shear control engineering, and relates to a hot-rolled strip steel optimized shearing method based on machine vision. The method is mainly combined with a machine vision principle, non-contact measurement is performed on a strip steel end deformation part caused by rough rolling on a hot rolling production line, optimization control is performed on a flying shear system through a fuzzy self-tuning PID controller, and the method comprises the following operation steps: 1) collecting a strip steel end image; 2) processing image information; 3) obtaining a three-dimensional contour; 4) performing three-dimensional convex hull operation; 5) making a fuzzy self-tuning PID controller decision; 6) transmitting a shearing signal by the controller; and (7) shearing by an executing mechanism. Based on a machine vision detection system, the three-dimensional profile of the end of strip steel is obtained in combination with an image processing unit, a flying shear mechanism is controlled through the fuzzy self-tuning PID controller, and a shearing mechanism carries out accurate shearing after receiving a signal sent by the controller. By means of a closed-loop control structure, online optimization of a hot-rolled strip steel flying shear system is achieved.

Description

technical field [0001] The invention belongs to the field of hot-rolled strip steel flying shear control engineering and relates to a machine vision-based optimal shearing method for hot-rolled strip steel. Background technique [0002] In the hot rolling production process, after the strip steel billet is rolled by the rough rolling unit, due to the uneven pressure of the rough roll or the influence of the environment, the end of the strip is prone to deformation. In order to ensure the stability of subsequent production, before entering the finish rolling unit Strip ends are required to be cut off. [0003] The traditional strip end shape detection uses offline and manual methods. The sampling time is long, the labor intensity is high, and the real-time information is poor. The detection cannot be effectively realized and the working efficiency of the rolling mill is reduced. At the same time, the flying shear system in the traditional hot rolling production line often ad...

AI Technical Summary

Problems solved by technology

[0003] The traditional strip end shape detection uses offline and manual methods, which takes a long time to sample, is labor intensive, and has poor real-time information, which cannot effectively realize the detection and reduce the working efficiency of the rolling mill.

At the same time, the flying shear system in the traditional hot rolling production line often adopts a fixed-length shearing system. Its inherent disadvantage is that the shearing length cannot be determined according to the shape of the head and tail of the strip, resulting in low shearing accuracy, and the shearing length varies from time to time.

In order to prevent dirty cutting, the operator often enlarges the cutting length artificially, which will easily cause an increase in the cutting loss and bring unnecessary waste

Images