The invention relates to a method and a system for processing thermoplastic resin and in particular method and system for processing thermoplastic resin using machine learning.
PET (Polyethylene Terephthalate) is now the industry material of choice for packaged high volume consumer products. Processors of PET experience property variations between suppliers' batches of virgin raw material. No solution exists to address quality variations in virgin PET raw materials. No liquid composition formula exists to facilitate the addition of PET re-cyclate (post-consumer/R-PET) into virgin raw material production stream.
Just as PET enclosures have mostly replaced glass bottles, similarly, spun PET fiber, produced by conventional methods using conventional machinery has largely replaced cotton in the insulation and fabric industry applications (clothing, upholstery, carpets, nappies).
PET extrusion conventional methods using conventional machinery is common methodology employed in all forms of food packaging (trays, containers).
PET injection molding conventional methods using conventional machinery is primarily used for making hallow mini-containers (pre-forms) which later are blow molded by heat/pressure to make bottles (for drinking liquids).
Due to raw material variations, PET processors experience between 4-6% non-conforming products (scrap). The accepted industry real cost calculation is generally 5 times the cost of this rejected processed material. Dry additive manufactures supply to compounders who typically only supply large batch quantities of virgin polymer (minimum 25 tons lots).
Up until now, only additives (such as “Repi™”, “Color Matrix™”) in the PET raw material processing process where property tuning additives for influencing such properties as color and clarity of processed material. But none of known additives where suitable for reducing or eliminating raw material variation in PET raw material processing.
Liquid additives are limited to changing certain properties of material and are used as colorants, UV stabilizers, anti-static additives, optical brighteners, temperature tolerance additives (to prevent damage in later processes), gas controllers e.g. oxygen scavengers, CO2 barriers, acetaldehyde scavengers etc. (to increase shelf life of finished products, preserve the taste and appearance of packaged products etc.). Theses additives are mostly aesthetic enhancers, but are not suitable for R-PET valorization, scrap reduction, reactive/closed loop.
Non-reactive processing systems have no means of rescuing a faulty of unevenly-mixed dry polymer blend once it is being put through molding, spinning, or extrusion equipment.
Just 2% of plastics are used in closed-loop recycling, where it is reintroduced to the value chain multiple times. This is just 1.6 million tones out of 78 million tones annual demand in Europe. In comparison, about 72% of four plastic demand goes completely unrecoverable.
Industry codes and pending legislation throughout Europe are focusing on packaging products made from PET. This is driving a more proactive approach to recycling and reducing carbon footprint in PET Processing Facilities.
The invention addresses the mentioned problems by overcoming said shortcomings and shows further advantages.
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.
In particular the invention allows (with respect to fiber spinning, extruding and inject blow molding) reduce scrap loss rate, rework, improve production yields, reduce cycle times, reduce pollution, reduce carbon footprint, reduce energy consumption, stabilize processing, reduce downtime, improve R-PET behavior in processing, increase overall R-PET utility.
The invention allows reduction of component scrap levels by at least 75% i.e. 4% scrap reduced to 1%, increase PET extrusion, PET injection molding, PET fiber spinning lines productivity.
Features and advantages of the invention are described in detail with reference to the drawing:
It should be understood that numerous specific details are presented in order to provide a complete and comprehensible description of the invention embodiment. However, the person skilled in art will understand that the embodiment examples do not limit the application of the invention which can be implemented without these specific instructions. Well known methods, procedures and components have not been described in detail for the embodiment not to be misleading. Furthermore, this description should not be considered to be constraining the invention to given embodiment examples but only as one of possible implementations of the invention.
According to first aspect of the invention a system for processing PET raw material or recycled PET material is disclosed. The system comprises any conventional plastic processing machinery, such as plastic spinning or molding, or extrusion machinery for PET spinning, molding or extrusion. It further comprises mechanical-hydraulic system for the on-line administration of liquid plastic additives. The additives improve behavior of PET in-process, particularly recycled PET (R-PET). The system automatically and intelligently modulates additive mixing ratios in plastic batches. The injection system comprises a carousel injection system for adding a range of property enhancers or property altering additives to a blend while it's being melted, molded or extruded. The system comprises a dynamic PID system comprising a programmable logic controller (PLC) and system of plurality of sensors. The sensors collects input from three key positions in any plastic processing machine—the material hopper, the screw, and the dispensing throat. Guided by PLC, which calculates chemical additive output based on its programming, appropriate liquid additive is then selected from a carousel and dispensed through a pump and manifold, injecting the additive into the liquid PET. Tangible systems are tied together with cloud connectivity. Data is access is available through internet connection. Allows remote monitoring and emergency response and a more holistic understanding of the system.
The PLC integrates sensors and control switches for system-critical hardware as well as the central processing and process safety hardware. The system comprises and integrated human-machine interface hardware.
According to second aspect of the invention a liquid composition for use in the system above is and its manufacturing method is disclosed. In particular liquid polymer additives and method of converting “dry” polymer additive compounds into “liquid” format is disclosed. The additive chemicals are based on oligomeric multifunctional materials dispersed in a liquid carrier.
According to third aspect of the invention a method for processing raw PET material or recycled PET material is disclosed. The method comprises steps:
The modulation of polymer being processed is based on known inputs and measured outputs at key stages in the plastic processing system—the dry raw material hopper, the extrusion/molding screw and the dispensing throat. The system is reactive and adaptable, responding to live data collected from machinery on the manufacturing line. This allows to respond to any aberrant behavior from the raw materials or dry blend. Continuous monitoring comprises automated sensor reading, automatic process capability and process capability indexing monitoring.
Software monitors and processes inputs from the sensors and manages the outputs of the carousel and injection systems. It dynamically responds to the amount of additive needed (proportional control), adjusting the injection rate and amounts as properties approach ideal behavior (differential control), and predicting behavior based on these inputs to avoid overcompensating and necessitating more additive (integral control).
The liquid additives are used to dynamically dose a shot of PET or R-PET based on data from relevant sensors. The dosing unit then delivers a specific small volume of these process modifiers based on sensor inputs, as calculated by the software. The chemicals then control the behavior of the plastic through heat-activated crosslinking or chain extension of the polymeric material during the melt processing step.
Adequate distribution of denial of service (DDoS) protection is present.
The process monitoring system collects key data for mixing ratios, temperature, injection screw force, material viscosity and other critical inputs. Based on the collected data the control system calculates the additive amounts needed based on desired results and where the inputs stand relative to these known values. From there it relies on fundamental control system principles of PID process control (proportional-integral-differential). This ensures close adherence to the desired outcomes, ensuring predictable system and output material behavior.
By improving thermal resistance in the blend prevents liquid shot from accidentally reaching degradation temperature due to unnecessarily high viscosity keeping the same section of the material in contact with heating coils for too long.
It is to be understood that the liquid agent introduction system according to the invention may be used in conjunction with polymer processing systems such as continuous and discontinuous systems and in particular blow molding, injection molding and melt spinning systems.
Although numerous characteristics and advantages together with structural details and features have been listed in the present description of invention, the description is provided as an example fulfilment of the invention. Without departing from the principles of the invention, there may be changes in the details, especially in the form, size and layout, in accordance with most widely understood meanings of the concepts and definitions used in claims.
Number | Date | Country | Kind |
---|---|---|---|
LT2017540 | Dec 2017 | LT | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/IB2018/050751 | 2/7/2018 | WO | 00 |