Texturing Process Control Network for Real-Time Yarn Quality Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The texturing process for Pre-Oriented Yarn (POY) to Draw Textured Yarn (DTY) is complex, with parameter settings significantly impacting the quality of the final yarn spindle, and existing methods struggle to optimize these parameters in real-time to prevent defects and improve product quality and efficiency.
Innovation Solution
A control method involving a parameter prediction model to forecast future control parameters, an index prediction model to assess and adjust parameters based on historical data, and a data generation model to generate desired parameters when deviations are detected, utilizing a control network with subnetworks for training and prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If real-time parameter prediction and adjustment is implemented, then product quality is improved, but system complexity increases
Solution Approach 1:
The control system is segmented into three specialized subnetworks: parameter prediction subnetwork (predicts future control parameters), index prediction subnetwork (predicts quality indices), and data generation subnetwork (generates optimized control parameters). Each subnetwork handles a specific function, reducing overall system complexity through modular design while achieving comprehensive quality control.
Solution Approach 2:
The parameter prediction subnetwork forecasts future control parameters before they are actually needed in the texturing process. This preliminary prediction allows the system to proactively adjust parameters to prevent quality deviations rather than reactively correcting them, improving yarn quality through advance planning.
2Productivity
If multiple prediction models are used to optimize parameters, then production efficiency is improved, but computational resources increase
Solution Approach 1:
The three subnetworks are merged into a single integrated control network that shares common infrastructure (data input layers, processing mechanisms, and output interfaces). This unified structure allows the system to leverage multiple prediction models for comprehensive optimization while avoiding redundant computational overhead through shared resources.
Solution Approach 2:
The data generation subnetwork automatically generates optimized control parameters based on predictions from the other subnetworks, eliminating the need for manual parameter tuning or external optimization tools. This self-service capability streamlines the optimization process and reduces overall computational resource requirements.
Data Source
AI summary
Provided is a control method for a texturing process, a training method for a control network and related apparatuses. The control method includes: sequentially collecting preset control parameters of a plurality of moments in the texturing process to obtain a first sequence; inputting the first sequence into a parameter prediction model to predict prediction parameters of a plurality of future moments; selecting a target parameter of a target moment from the prediction parameters; building a second sequence containing the target parameter based on the control parameters before and after the target moment; processing the second sequence based on an index prediction model to obtain an index prediction result; and inputting the result into a data generation model in a case where the result does not meet a desired value, to obtain a desired control parameter for the target moment.