Longitudinal Crack Prediction in Steel Continuous Casting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Previous methods for predicting longitudinal cracks in continuous steel casting using direct temperature measurement are unreliable due to high failure rates of thermal elements and poor connections, leading to inconsistent signal quality.
Innovation Solution
A method involving multiple rows of thermal elements in the mold wall for measuring local strand temperature, with a statistical assessment using principal component analysis (PCA) and an expert decision system to correct signals based on spacing and strand speed, and defining risk factors for countermeasures such as adjusting casting speed or electromagnetic brakes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If thermal elements are arranged in multiple rows distributed along the mold height for temperature measurement, then measurement precision and predictive accuracy improve, but device complexity increases
Solution Approach 1:
The mold wall is segmented into multiple rows of thermal elements distributed along the height, with each row providing independent temperature measurement capability. This segmentation enables precise localization of temperature anomalies and improves detection accuracy without requiring a single complex sensor system
Solution Approach 2:
The thermal elements are arranged in a two-dimensional distribution pattern (multiple rows along the height and across the width of the mold), transforming a one-dimensional measurement approach into a two-dimensional temperature field mapping system. This dimensional expansion provides comprehensive coverage and enables precise crack prediction through spatial temperature analysis
2Loss of information
If thermal elements are used for direct temperature measurement, then temperature data can be obtained, but reliability deteriorates due to high failure rates and poor connections
Solution Approach 1:
Multiple thermal element signals are merged and evaluated collectively through statistical methods rather than relying on individual element reliability. The system combines data from all rows and thermal elements to produce a robust temperature assessment that compensates for individual element failures or connection issues
Solution Approach 2:
The system continuously monitors temperature signals from all thermal elements and uses feedback loops to detect anomalies, correct signals, and update predictions. When connection problems or failures occur, the feedback mechanism identifies affected elements and adjusts the evaluation based on remaining functional elements and historical patterns
3Measurement precision
If statistical assessment with multiple thermal element rows is implemented, then predictive accuracy for longitudinal cracks improves, but loss of time increases due to signal correction and timing alignment requirements
Solution Approach 1:
Temperature data from all thermal element rows is collected and stored in advance before crack prediction is required. The system maintains a historical record of temperature signals and performs preliminary processing, so when crack prediction is needed, the analysis can be performed quickly using pre-organized data
Solution Approach 2:
The system transforms raw temperature signals into corrected temperature values by applying timing corrections based on strand position and speed. This parameter transformation converts complex multi-row data into a standardized format that can be rapidly evaluated for crack prediction without requiring extensive real-time processing
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 approach significantly improves predictive accuracy for longitudinal cracks by correcting thermal element signals and implementing timely countermeasures, reducing the risk of crack occurrence through precise temperature gradient analysis.
Implementation Method 1
the local strand temperature is measured by thermal elements arranged so as to be distributed in a mold wall
Data Source
AI summary
A process for predicting longitudinal cracks during continuous casting of steel slabs. The local strand temperature is measured by thermocouples distributed in the mold wall. In this process, the risk of the strand rupturing as a result of longitudinal cracking is assessed statistically taking into account the current temperature values measured by the thermocouples arranged in the mold and the temperature values determined when no cracks are present.