PET Processing with Real-Time Additive Dosing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The processing of PET raw materials results in significant property variations and high scrap rates due to raw material inconsistencies, with existing methods failing to effectively manage or reduce these issues, particularly in recycling and re-introduction of post-consumer PET into production streams.
Innovation Solution
Employing real-time process management and machine learning to dynamically adjust the dosing of homogenizing composition and chain extenders using liquid additives as carriers, allowing for continuous monitoring and adaptive control of PET processing, including the use of a mechanical-hydraulic system with sensors and a programmable logic controller to optimize additive mixing ratios and material performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional PET processing methods are used, then production efficiency is maintained, but scrap rates increase to 4-6% due to raw material variations
Solution Approach 1:
The patent implements dynamic adjustment of processing parameters (temperature, pressure, additive dosing) based on real-time monitoring of raw material properties. The system adapts processing conditions to match variations in PET batch quality, preventing scrap formation while maintaining continuous production efficiency.
Solution Approach 2:
The system incorporates real-time monitoring and feedback mechanisms that continuously measure material properties during processing and automatically adjust parameters to compensate for variations. This closed-loop control prevents defect formation and reduces scrap rates while maintaining productivity.
2Loss of substance
If real-time process management and machine learning are implemented, then scrap rates reduce by 75%, but device complexity increases
Solution Approach 1:
The system employs machine learning algorithms that automatically learn optimal processing parameters from historical data and real-time measurements without requiring manual intervention. The system self-adjusts to material variations, reducing scrap rates while the automation masks the underlying complexity from operators.
Solution Approach 2:
The patent replaces manual process adjustment and traditional control systems with automated machine learning-based control. This substitution of mechanical/manual operations with intelligent algorithms reduces scrap through precise parameter optimization while the complexity is contained within the software layer rather than requiring complex physical systems.
3Manufacturing precision
If liquid additives are used for property enhancement, then material performance improves, but ability to rescue faulty blends is lost
Solution Approach 1:
The system dynamically adjusts additive dosing rates and types based on real-time material property measurements. When faults or variations are detected, the system can modify the composition and amount of liquid additives in real-time to compensate for deviations, maintaining both performance precision and adaptability to faulty blends.
Solution Approach 2:
The patent changes the physical or chemical parameters of the liquid additives (concentration, type, dosing rate) in response to real-time material conditions. This allows the system to optimize material performance while simultaneously adapting to and correcting faulty blends through parameter modification rather than fixed additive formulations.
Data Source
Figure 1
AI summary
The invention addresses the above problems by tuning PET raw material processing process by employing real time process management and machine learning steps and reactive addition of dosing of homogenizing composition for impregnating chain extenders and compatibilizing agents in thermoplastic resin using the liquid additive as a carrier into the process for modifying material performance. The properties of PET blend are no longer fixed once dry-blending and melting is complete.