Surface Defect Detection on Continuously Cast Slabs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting and eliminating surface defects on continuously cast products are not reliable, as they often remove unnecessary flaws and depend on stored models, leading to inefficiencies and subjective errors in evaluating and classifying defects.
Innovation Solution
A method that uses topographical information to detect and classify surface defects on continuously cast products by determining their exact position and dimensions, differentiating between irrelevant flaws and relevant defects using classification methods like neural networks, and linking inspection data from slabs to finished products for improved evaluation and elimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all surface flaws are eliminated, then product quality is improved, but unnecessary work steps are performed and productivity decreases
Solution Approach 1:
The system performs preliminary actions by detecting and storing defect information from the slab surface before rolling, then uses this information to guide selective elimination. The inspection device captures topographical data in advance, creating a database that informs subsequent selective grinding operations, avoiding the need to eliminate all flaws uniformly
Solution Approach 2:
The system applies local quality by treating different surface defects differently based on their specific characteristics and locations. Instead of uniform elimination, the system selectively eliminates only those defects that will actually cause problems in the finished product, while leaving benign defects untouched. This is achieved through comparing slab inspection data with finished product inspection data to identify which defects are truly harmful
2Device complexity
If stored models are used for defect evaluation, then classification is simplified, but subjective errors and unreliable evaluation occur
Solution Approach 1:
The system implements feedback by using actual inspection results from the finished product to continuously improve and update the classification models. The comparison between slab surface defects and finished product defects provides feedback that refines the understanding of which defects are truly harmful, reducing subjective errors over time through iterative learning
Solution Approach 2:
The system performs self-service by automatically comparing defect data from slab inspection with finished product inspection results, enabling it to self-correct and improve its classification accuracy without external intervention. The system learns from its own operational data to refine which defects require elimination
3Reliability
If comprehensive defect detection is performed, then detection reliability is improved, but measurement and evaluation complexity increases
Solution Approach 1:
The system extracts only the relevant information needed for decision-making by comparing comprehensive defect detection data with finished product inspection results. It separates harmful defects from benign ones by extracting the subset of defects that actually correlate with finished product quality issues, simplifying the evaluation process while maintaining detection reliability
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method ensures more reliable detection and elimination of only those defects that impact the quality of the finished product, reducing unnecessary work steps and subjective errors through a self-learning process that predicts defect occurrence based on topographical data.
Implementation Method 1
In the visible range, fringe projection methods and stereoscopic methods are used
Implementation Method 2
In the visible range, fringe projection methods and stereoscopic methods are used
Implementation Method 3
Laser-based methods are also possible
Data Source
AI summary
A method for the detection and classification of surface defects on continuously cast products using topographical information about the appearance of continuously cast surface defects and/or flaws are determined with respect to their exact position, evaluated with respect to their location and dimensions, and eliminated in accordance with the evaluation prior to further machining of the product, or are prevented by optimizing the process. The defects and/or flaws on the slab surface of the continuously cast preliminary product are detected and are stored with respect to their exact position and a detection of defects and/or flaws on the finished product is carried out and stored with respect to their exact position, and in that the information from the preliminary product is then compared with the information from the surface inspection on the finished product. Only the information which has led to, or can lead to, defects on the finished product is considered for the elimination of defects and/or flaws on the preliminary product.