Air Knife Coating Weight Control Using Neural Network Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling coating weight in hot dipping processes face limitations in accuracy and surface quality due to manual operation and difficulties in deriving optimal air knife gap and pressure settings for various steel types and conditions.
Innovation Solution
A coating weight control apparatus and method utilizing a prediction model trained with neural networks to derive absolute values of air knife gap and pressure based on input operation conditions, incorporating statistical methods and look-up tables to improve control accuracy and surface quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual operation with reference to operator experiences or setting tables is used to adjust air knife gap and pressure, then the control method is simple to implement, but the coating weight control accuracy is insufficient and surface quality deteriorates
Solution Approach 1:
The patent replaces manual mechanical adjustment of air knife gap and pressure with an automated control system using neural networks and prediction models. The system automatically derives optimal air knife gap and pressure values based on steel sheet information, eliminating manual operation while improving coating weight control accuracy to within ±2 g/m².
Solution Approach 2:
The patent changes the control parameters from manual experience-based adjustments to scientifically derived parameters using neural networks. The system calculates optimal air knife gap (in mm) and air knife pressure (in kgf/cm²) based on steel sheet thickness, width, and coating weight requirements, transforming qualitative manual control into quantitative automated control.
2Measurement precision
If coating weight gauge is positioned at rear end 200m from air knife to measure dried plating layer, then measurement is accurate, but feedback control cannot be immediate and response time increases
Solution Approach 1:
The patent performs preliminary calculation of optimal air knife gap and pressure before the steel sheet reaches the air knife position. The neural network prediction model computes the required parameters in advance based on steel sheet information, enabling proactive control rather than reactive adjustment after coating deviation occurs.
Solution Approach 2:
The patent introduces an intermediate prediction model and neural network calculation system between the steel sheet information and the air knife control. This intermediary system processes the control logic remotely, allowing accurate prediction and control parameter derivation without requiring physical proximity between measurement and control points.
3Adaptability or versatility
If air knife gap and pressure are adjusted based on statistical methods or look-up tables, then the control method is easier to implement, but adaptability to various steel types and conditions is limited
Solution Approach 1:
The patent replaces static statistical methods and look-up tables with a dynamic neural network prediction model. The neural network automatically adapts to different steel types, thicknesses, and coating requirements by learning from training data, providing high adaptability without requiring manual intervention or pre-programmed tables for each condition.
Solution Approach 2:
The patent transforms the control system from static (fixed statistical methods or look-up tables) to dynamic (neural network-based real-time prediction). The system continuously adapts its predictions based on input steel sheet parameters, enabling automatic adjustment to various steel types and coating conditions without manual reconfiguration.
Data Source
AI summary
Provided is a method of controlling coating weight coated on a strip by using an air knife disposed in a travelling direction of the strip in a continuous plating process in which the strip is dipped in a molten metal pot and is coated. The method includes: training a neural network with accumulated operation conditions; and deriving an absolute value of at least one of an air knife gap and an air knife pressure by using the trained neural network based on an input operation condition.