Polymer extrusion control with artificial intelligence techniques

Currently, polymeric materials play a major role in production industry but require more
advanced process control to improve product quality and process efficiency. The control of melt
quality and throughput has traditionally been attempted by joint regulation of melt pressure and
melt temperature. The viscosity control has also been considered by a number of researches
whilst some work attempted alternative approaches (i.e., apart from the temperature, pressure
and viscosity) for process control. Overall, the majority of control schemes have been
concentrated on linear techniques which have demonstrated some potential to reduce
fluctuations in melt pressure, temperature and melt viscosity, but as extrusion is highly nonlinear,
they are not suitable for a wide operating range which applies in industrial environments. Some
alternative control techniques (e.g., Artificial Intelligence (AI)) are capable of handling process
nonlinearities and only a limited amount of work can be found in the literature based on these
techniques for barrel set temperature control within desired limits and did not consider about
melt temperature or melt quality control. Hence, this project aims to investigate an advanced
process control strategy which can effectively handle the process nonlinearities over a wide
operating window. The use of the techniques such as AI in the melt temperature control,
individually or jointly with melt pressure may be desirable as they can handle process
nonlinearities even without having exact numerical details of the process. Initially, the efficacy of
the novel controller should be explored via simulation and then should be tested on an extruder.
The required experimental studies can be performed on a medium scale extruder with widely
used polymeric materials. Ideally, the novel controller should facilitate to optimize the process
energy efficiency and product quality.