Machine Vision-Based Crystallizer Nozzle Centering for Continuous Casting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing crystallizer nozzles require manual intervention for centering, which is inaccurate due to potential blockages and inability to smoothly discharge solvent, affecting the uniformity and quality of the crystallization process.
Innovation Solution
A machine vision-based method using image acquisition, edge enhancement, ant algorithm for edge extraction, and morphological closing operations to determine centerline equations, enabling automated centering of crystallizer nozzles in both width and thickness directions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual intervention is used to adjust and monitor the crystallizer nozzle, then the system is simple to operate, but the centering accuracy is low due to potential blockages and inability to smoothly discharge solvent
Solution Approach 1:
The patent replaces the manual mechanical adjustment system with an automated machine vision system. Image acquisition apparatus captures images of the crystallizer nozzle, and image processing algorithms automatically determine the nozzle center position and generate adjustment instructions, eliminating manual intervention and achieving high-precision centering without human error or fatigue.
Solution Approach 2:
The system enables self-service by allowing the crystallizer nozzle centering to be performed automatically through image acquisition and processing. The nozzle itself provides the visual features needed for detection, and the system self-corrects positioning without requiring external manual adjustment, achieving autonomous operation.
2Productivity
If manual adjustment is used for crystallizer nozzle centering, then the device complexity is low, but the productivity and efficiency are reduced due to time-consuming manual operations
Solution Approach 1:
The patent replaces manual mechanical adjustment operations with an automated machine vision system. Image acquisition apparatus and image processing algorithms automatically detect nozzle position, calculate center coordinates, and generate adjustment instructions, significantly reducing centering time and improving productivity while eliminating repetitive manual labor.
Solution Approach 2:
The system enables continuous operation by automatically performing image acquisition, processing, and adjustment instruction generation without interruption. The machine vision system can continuously monitor and adjust the crystallizer nozzle position, maintaining optimal centering throughout the production process without manual intervention delays.
3Reliability
If manual monitoring is performed on crystallizer nozzle position, then the system is simple, but the reliability of centering is low due to potential blockages affecting observation
Solution Approach 1:
The patent replaces manual visual monitoring with an automated machine vision system. Image acquisition apparatus continuously captures images of the crystallizer nozzle, and image processing algorithms automatically analyze the images to determine nozzle position and center coordinates, providing reliable and consistent monitoring without human error or fatigue.
Solution Approach 2:
The system implements feedback by continuously acquiring images of the crystallizer nozzle, processing the images to determine position deviations from the center, and generating adjustment instructions based on the analysis results. This closed-loop feedback mechanism ensures reliable centering by automatically detecting and correcting position deviations.
Data Source
AI summary
A machine vision-based crystallizer nozzle centering method includes: S1, performing image acquisition on a tundish car lower nozzle and a crystallizer nozzle; S2, performing an edge enhancement using guide image filtering; S3, performing an edge extraction on a first image set using an ant algorithm, and obtaining a second image set; S4, performing smoothing processing and a centerline extraction on the second image set using a morphological closing operation respectively, and obtaining a centerline equation; S5, centering the crystallizer nozzle in a width direction according to the centerline equation, and obtaining a width centering result; S6, installing two second image acquisition apparatuses symmetrically on a center extension line of the crystallizer nozzle, and obtaining a thickness centering result using the second image acquisition apparatuses; and S7, adjusting a position of a tundish car according to the width centering result and the thickness centering result.