Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.554339
Title: Modelling and control of melt temperature in polymer extrusion
Author: Abeykoon, Yapa Mudiyanselage Chamil Eranda Kumara
Awarding Body: Queen's University Belfast
Current Institution: Queen's University Belfast
Date of Award: 2011
Availability of Full Text:
Full text unavailable from EThOS. Please contact the current institution’s library for further details.
Abstract:
The extrusion process serves as one of the main methods for processing polymer materials and thermal homogeneity of the process output presents a major challenge for high quality extruded products. Despite considerable research, process monitoring and control still remains difficult and engineers have to deal with issues such as the selection of process settings and product quality control primarily by trial and error. In this study, it was recognised that melt temperature was the most appropriate parameter for real-time investigation of process melt quality and a new strategy for polymer extrusion process control incorporating an inferential thermal monitoring technique was examined. An experimental evaluation showed that existing point/bulk measurement techniques were poor at capturing in-process thermal fluctuations although they are common in practice and therefore a multi point thermocouple mesh technique was selected to use in the experiments. As the thermocouple mesh was not yet robust enough for industrial applications, an inferential thermal monitoring technique was explored to use in the prospective process controller. Initially, attempts were made to inferentially predict the process thermal stability via screw load torque and melt pressure dynamics and these were not successful. Therefore, the experimental data collected from the thermocouple mesh was used to develop a dynamic model to predict the temperature profile of the extruder output melt flow and this model was used to develop a soft sensor for real-time prediction of a melt temperature profile. The possible prediction errors of the soft sensor were corrected by a feedback model engaged with a physical infrared temperature sensor measurement. Simulation results showed that the soft sensor is capable of predicting melt temperatures at different radial locations of the melt flow with good accuracy and the feedback model prediction error was always less than ±2.25% of the full scale reading; hence this provided a promising approach for inferential process thermal monitoring. Ultimately, a model-based control approach incorporating the soft sensor and fuzzy logic was proposed to minimise melt temperature variance across the extruder output melt flow while achieving the desired average melt temperature by manipulating the major process variables (i.e. screw speed and all the barrel zone temperatures). Simulation results showed that the proposed controller was good in achieving its targets over the experimental results. Therefore, this offers a new method to operate extruders at high screw speeds whilst achieving both high energy and thermal efficiencies. The proposed control framework is industrially usable with the development of the required process models as it uses only readily measurable process parameters for its operation. Presently, the controller's operation is specific to a given machine and material and hence the development of generalised models should help to widen its scope. Although the single screw extruders were used for research, the proposed methodology should be applicable to other multi screw extruders as well where the development of the required models is possible.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.554339  DOI: Not available
Share: