Electrical & Electronic Engineering - Stellenbosch University

Home
South Africa
Stellenbosch
Electrical & Electronic Engineering - Stellenbosch University

Directions Website

28/10/2026

How can batteries at different voltages share energy in both directions?

For his 2025 final-year project in Stellenbosch University's Department of Electrical and Electronic Engineering, Corben Smith designed and built a bidirectional power converter under the supervision of Dr F. Mwaniki.

Tests with a 72-volt electric motorcycle battery and a lower-voltage battery pack delivered approximately 96.2% efficiency in each direction. The measured output reached 965 watts in step-down mode and about 951 watts in step-up mode.

The tests also identified a development priority: step-up operation began oscillating above roughly 700 watts. Further controller tuning and protection measures are needed as the work progresses towards potential home energy applications.

Read the full article: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/design-and-implementation-low-voltage-bi-directional-dc-dc-converter

30/09/2026

Finding whale calls in days of underwater audio takes time, especially when background noise masks the sounds researchers want to study.

For his Electrical and Electronic Engineering project at Stellenbosch University, Jonathan Rule, under the supervision of Professor Jaco Versfeld, developed software to detect and classify Bryde’s whale vocalisations in recordings from False Bay. It combines audio processing with hidden Markov models, which use patterns over time to identify likely calls.

In a two-minute evaluation clip, the detection configuration found 28 of 30 labelled calls and produced six false alarms. A separate classification configuration assigned the correct call type to 26 calls, misclassified four, and produced seven false alarms. A detection counted as correct when it overlapped any part of a labelled call, so these figures don’t establish accurate call boundaries.

The software exports suggested labels to Raven Pro for researchers to review. Further testing on longer recordings and rare call types is needed to assess its use in broader monitoring.

Read the full article for the approach, findings, and limitations: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/automated-detection-and-classification-brydes-whale-vocalisations

29/09/2026

A wind forecast error can change which generators need to run and the cost of keeping electricity supply in balance with demand.

For his Electrical and Electronic Engineering project at Stellenbosch University, Oliver Jonas under supervision of Dr. Chantelle van Staden combined historical data from Sere Wind Farm with signal processing and machine learning to forecast wind power for the next 24 hours.

His thesis reports an average absolute forecast error of 9.84 megawatts for the final hybrid model, compared with 28.30 megawatts for a baseline using historical hourly averages.

He then tested one selected day’s forecast in a generator scheduling model representing three South African generation mixes. The simulations showed how remaining errors could increase reliance on gas generation and leave some renewable energy unused.

The findings depend on the study’s historical tests, single-farm data, and simplified scheduling model. Broader testing would help assess how the approach could support operational planning.

Read the full article for the methods, results, and limits of the study: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/wind-power-forecasting-and-uncertainty-informed-generator-scheduling-unit-commitment

31/08/2026

How can a superconducting qubit design move from chip geometry to a computable model of quantum behaviour?

In his engineering research project, Reece Bird unders supervision of Professor Coenrad Fourie, developed a software toolchain that automates this process. Starting with verified designs from the SQuADDS database, the program recreates each layout, prepares and runs InductEx and PALACE simulations, constructs Hamiltonians using two independent methods, and generates a detailed report.

Tests on five qubit-cavity designs found that the capacitances used in the calculations were within 10% of established reference values. PALACE eigenmode frequencies were within 7% of the reference simulations. Across the principal quantum parameters, about 85% of results were within 20%, and many were closer.

The comparison also quantified an important trade-off. The Lumped Oscillator Model averaged 20.6 minutes, while the more general Energy Participation Ratio method averaged 92.6 minutes and produced more consistent results. Two layouts also showed why direct eigenmode results remain preferable for cavity frequency under some geometric conditions.

Read the full article: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/chip-geometry-quantum-behaviour-automating-superconducting-qubit-analysis

28/08/2026

In celebration of Women’s Month, some of the women who continually support the Department of Electrical and Electronic Engineering were treated to a special lunch in recognition of their kindness, dedication and the important work they do. 🌸

On Wednesday, 26 August, the group gathered at The Pottery in Stellenbosch for lunch and pottery painting, while taking time to connect and celebrate.

Representing our academic, administrative and technical staff, as well as departmental assistants, these women each play a valuable role in our Department.

Thank you to every woman who contributes her knowledge, skill and dedication to our EE community.

26/08/2026

How can a high-altitude UAV regain its position when satellite navigation is disrupted?

In his Electrical and Electronic Engineering research project, William Kevin Morley, under supervision of Professor Rensu Theart, developed a lightweight visual place recognition system that allows an unmanned aerial vehicle to compare its current camera view with a sparse trail of geo-tagged ‘breadcrumbs’ captured earlier in the flight.

On the high-altitude test route, the system used only 26 stored breadcrumbs to accept 1,198 loop closures. Every accepted match identified the best available breadcrumb, while the storage method reduced retained data by 98.91%. A full recognition cycle took about 87.5 milliseconds on the CPUs used for testing.

The research also identified a clear limit. The system accepted no matches over lower, visually repetitive terrain where the outbound and return routes were widely separated. This contrast shows why terrain, viewpoint, sensor quality, and matching thresholds all need to be considered when developing navigation systems for disrupted GNSS conditions.

Read more about the project and its findings in the full article: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/helping-uavs-find-their-way-when-satellite-navigation-fails

28/07/2026

A final-year mechatronic engineering project at Stellenbosch University’s Faculty of Engineering explored how raw radar reflections can be reconstructed into a two-dimensional image.

Satheesh Bissessur, under the supervision of Mr Lanche Grootboom, developed a MATLAB algorithm that processes measurements recorded while targets rotate through 360 degrees in front of a radar system. The project combined background subtraction, windowing, range alignment, back projection, and minimum-entropy focusing.

The final algorithm located both single and multiple targets and achieved a measured resolution of 1.65 cm in the distance and cross-range directions. It also showed how reducing the number of viewing angles affects image focus and spatial detail.

The project provides a foundation for future work on more complex target shapes, improved positional accuracy, and other radar-imaging applications.

Read the full article: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/radar-reflections-images-reconstructing-rotating-scenes-isar

21/07/2026

When radio signals don't behave as planned at the Square Kilometre Array site in South Africa, the cost isn't just technical, it's time and money spent on redesign.

A final-year engineering project at Stellenbosch University by Me Nikita van Rooi, supervised by Prof Riaan Wolhuter, compared two radio propagation models across 411 site links to understand exactly when and why each model fails.

The result: ITU-R P.526 outperformed Longley-Rice, with 78% accurate predictions versus 68%. Machine learning identified the specific terrain features — clearance ratio, obstacle height, mean elevation — that drive failures in each model.

A practical framework that engineers can use to flag high-risk links before they become costly problems.

Read the full article: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/comparison-radio-propagation-models-receive-level-anomalies-ska

30/06/2026

How can a target impact be confirmed when visual spotting becomes difficult over long distance?

A final-year mechatronic engineering project at Stellenbosch University explored that question by developing a wireless Target Hit Indicator for steel gong targets.

Built by Dean van Zyl, the prototype combines accelerometer-based impact detection with LoRa radio communication. When the target is struck, the transmitter detects the impact and sends a feedback packet to a handheld receiver with an OLED display.

Field testing showed reliable communication at 1 km, 206.8 ms feedback latency, 100% impact detection accuracy in the test setup, and an estimated battery life of more than 100 hours.

The project brings together sensor selection, PCB design, firmware development, radio testing, and practical prototype validation.

Read the full post: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/turning-target-impacts-instant-wireless-feedback

11/06/2026

A final-year engineering project at Stellenbosch University explored how a locally hosted Home Energy Management System can help households make better use of solar, battery storage, and outage planning.

Built by Aldrich Opperman, under the supervision of Dr Arnold Rix, the system runs on a Raspberry Pi with a touchscreen dashboard that brings together inverter data, weather forecasts, load-shedding schedules, and smart-switch control.

The project moves past simple monitoring. It predicts household demand, estimates solar generation, and adjusts battery charging and selected loads ahead of outages.

It’s a practical piece of engineering shaped by South Africa’s energy context, and by the day-to-day decisions households face when grid supply is uncertain.

Read the full post: https://www.su.ac.za/en/faculties/engineering/departments/electrical-electronic-engineering/news/smarter-way-manage-home-energy-during-load-shedding

Address

Stellenbosch
7600

Website

http://www.ee.sun.ac.za/

Alerts

Be the first to know and let us send you an email when Electrical & Electronic Engineering - Stellenbosch University posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.