Wits’ virtual lab streamed live bioethanol fermentation data via MATLAB® and ThingSpeak™, giving 100+ students real-time process control experience and industry-ready skills.
CLIENT
The University of the Witwatersrand
sector
Education
Read time
10 Min
Overview
The Challenge
The Solution
The Results
The numbers
The Challenge
Professor Antony Higginson had been using the Wits Microbrewery as part of his Ph.D. research into producing bioethanol, a renewable fuel that can be blended with gasoline or used directly. He saw an opportunity to bring this real-world process into his teaching to give students a richer learning experience.
However, several barriers stood in the way:
Large Class Sizes: Over 100 fourth-year chemical and metallurgical engineering students, making in-person lab work impractical.
Limited Resources & Space: The brewery couldn’t accommodate all students, and industrial SCADA systems are costly to build and maintain.
COVID & Remote Teaching Constraints: Students needed an engaging, hands-on experience that could work virtually.
Access & Infrastructure Challenges: Unstable electricity supply in South Africa (load shedding) could disrupt server access, and students had varying levels of device and internet availability.
Higginson wanted to create a setup that would deliver authentic, real-time process control experience to all students without expanding the physical lab footprint.
The Solution
Higginson integrated his bioethanol fermentation experiments into the Process Control course, creating a virtual lab experience:
Live Data Streaming: Sensors measuring variables such as temperature and density fed into a PLC, two Arduino® microcontrollers, and a Raspberry Pi® server. Data was streamed to ThingSpeak™ via Wits’ MATLAB® Campus-Wide License, allowing students to monitor experiments in real time from any device, anywhere.
Webcam Access: Students could watch Higginson set up equipment and follow the process remotely.
Computational Modelling: Students built simplified MATLAB models grounded in physics principles to replicate brewery data. They learned to fit key parameters like heat transfer between system elements and pump energy input.
Real-Time Problem Solving: Students selected which data to use, excluding anomalies such as when a cooling pipe iced over, halting heat transfer.
Industry Simulation: The setup mimicked industrial SCADA systems at a fraction of the cost, giving students exposure to operational control environments.
Resilience to Load Shedding: Cloud-hosted ThingSpeak ensured data access even during power outages, as long as students had a battery-powered device and cellular internet.
Scalable & Shareable: The setup could be shared across universities, enabling collaborative experiments similar to other cloud-connected engineering labs presented at the IFAC Symposium on Advances in Control Education.
The microbrewery plant at the University of the Witwatersrand, showing the fermentation vessel on the right. (Image credit: Antony Higginson)
The Outcome
Enhanced Learning Experience: Students gained hands-on experience with real, messy industrial data, improving skills in process control, data analysis, and engineering tools.
Graduate Attribute Alignment: Met the Engineering Council of South Africa’s requirement for assessing the use of engineering tools and information technology.
Skill Application Beyond University: Graduates applied MATLAB and IoT skills in diverse industries, from pulp and paper to minerals processing and craft distilling.
Scalable Education Model: Demonstrated that a virtual lab can scale beyond 100 students and be replicated by other institutions.
Research Synergy: The same system supported Higginson’s AI research, using neural networks to optimize ethanol yields.
Operational Efficiency: Eliminated the need for manual data collection from SCADA PCs; cloud integration made collaboration with supervisors and peers seamless.
“This project is about giving the students an experience of a system like SCADA without having to greatly expand the amount of available lab equipment.” – Prof Antony Higginson
ThingSpeak displays the raw live data from each sensor. For more data visualization, visit here. (Image credit: Antony Higginson)
.png)
.png)
