Faculty/School

Faculty of Engineering

School of Mechanical, Medical and Process Engineering

Topic status

We're looking for students to study this topic.

Research centre

Centre for Agriculture and the Bioeconomy

Primary Supervisor

Professor Azharul Karim
Position
Professor
Division / Faculty
Faculty of Engineering

Other QUT supervisors

Dr Zachary Welsh
Position
Lecturer
Division / Faculty
Faculty of Engineering

Overview

This project extends a recently completed thesis that developed a fuzzy logic controller (FLC) for regulating magnetron power in hybrid microwave-convective drying, validated through a MATLAB-COMSOL co-simulation framework. That work demonstrated successful single- and multi-sample (27-sample) temperature control with strong spatial uniformity, but was limited to a single-magnetron, lab-scale geometry.

The project builds on an existing industry collaboration between the ADSER group and industry partner, whose multi-magnetron industrial hybrid dryer is the intended target application for the framework.

Research engagement

The student will engage with:

  • Computational/simulation-based work, extending the existing MATLAB-COMSOL LiveLink co-simulation environment to a multi-magnetron cavity geometry.
  • Control system design, developing and tuning individual magnetron-level FLC agents plus a supervisory coordination agent.
  • Model validation and benchmarking, comparing the multi-agent controller against fixed-parameter operation and against the existing single-magnetron FLC results.
  • Optionally, stakeholder/industry engagement, incorporating feedback from industry on practical constraints of the industrial dryer.

Research activities

The student will:

  • Extend the existing COMSOL multiphysics model to represent a multi-magnetron cavity, building on the validated single-magnetron geometry and mesh.
  • Design and implement a multi-agent control architecture in MATLAB, with each magnetron operating its own FLC informed by feedback from the sample(s) most influenced by that magnetron, coordinated by a supervisory agent.
  • Run simulation studies comparing multi-agent control performance (temperature regulation, spatial uniformity/COV, drying rate) against fixed-intermittency operation and the prior single-magnetron benchmark.
  • Document and present findings, contributing toward a thesis and potentially a co-authored publication.

Research skills

The student will gain skills in:

  • MATLAB-based control system design, including fuzzy logic control and multi-agent coordination.
  • Co-simulation workflow development (MATLAB-COMSOL LiveLink).
  • Critical evaluation of control strategies against performance metrics (temperature overshoot, coefficient of variation, drying kinetics).
  • Technical writing and scientific communication, including engagement with an industry partner.
  • Broader systems-thinking skills applicable to industrial process control problems beyond food drying.

Outcomes

The project aims to:

  • Demonstrate whether a multi-agent FLC architecture can maintain the temperature control and spatial uniformity performance previously achieved at single-magnetron scale, when extended to a multi-magnetron industrial configuration.
  • Quantify any trade-offs introduced by scaling (e.g., inter-magnetron interaction effects, uncoordinated vs coordinated agent performance).
  • Produce a validated simulation framework that can directly inform control system design for iCCon Solutions' industrial hybrid dryer.
  • Generate results suitable for a thesis chapter and potential conference/journal publication, and lay groundwork for future experimental validation on the physical multi-magnetron system.

Skills and experience

nil

Start date

2 November, 2026

End date

19 February, 2027

Location

QUT Gardens point

Contact

Azharul Karim

0422343395

azharul.karim@qut.edu.au