Browse
Recent Submissions
Item The value of demand response in South Africa's electric power system(2025) Mdhluli, Sipho D; Makopo, Raisibe S; Mukoma, Peter; Madhoo, H; Figlan, TThe increasing adoption of grid-connected Distributed Energy Resources provided Distribution (and Transmission) Network Operators with the opportunity to procure a range of "flexibility services" from these third partyoperated assets in a way that is beneficial to the local or national electricity network—examples of flexibility services that can be procured include large-scale, grid-connected battery storage systems and demand response. This paper will explore different types of demand response interventions utilised globally, provide a South African context of the current demand response, and, lastly, use Plexos to quantify the system value that untapped demand response potential can provide to the South African electricity markets if pursued.Item Predictive analytics for proactive email security risk management: A systematic review(2024-12) Singano, Zothile T; Mudau, Tshimangadzo C; Ngejane, Hombakazi C; Ndlovu, Lungisani; Mncwango, Lungisani SIn recent years, email has become an important communication tool for sharing private messages to crucial business message exchanges. However, its widespread use makes it a major target for cyber-attacks, including phishing, spam, and malware. These growing threats highlight the urgent need to investigate email security risk management to protect against attacks and maintain the integrity of communication systems. The study reviews the literature on the challenges of email security, risk management, and the role of predictive analysis in combating these threats. Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), thirty-five (35) relevant peer-reviewed research articles were identified in various open research databases. This systematic literature review (SLR) also includes relevant case studies. The findings reveal that the integration of machine learning (ML), natural language processing (NLP), and real-time data analytics into email security frameworks improves threat detection and mitigation. Furthermore, these models often lack adaptability across languages and cultures. Additionally, they do not integrate well with human-centric security measures. Therefore, it is important to develop culturally adaptive predictive models, sector-specific solutions for industries such as finance and healthcare and incorporate behavioural analytics to enhance email threat detection and prevention. In other words, a comprehensive approach that combines technical advances with behavioural insights is crucial to strengthening email security and maintaining the integrity of global digital communications amid evolving cyber threats.Item Review of Machine Learning Techniques That Enable Network Slicing in Open RAN(2025) Mthethwa, Nosipho BP; Mwangama, J; Masonta, Moshe TThe evolution of 5G networks and the anticipated capabilities of 6G have placed significant emphasis on intelligent Radio Access Network (RAN) slicing to meet the diverse and stringent requirements of services such as enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communications (mMTC). This paper reviews state-of-the-art techniques, with a particular focus on the application of machine learning (ML) approaches in RAN slicing. The study categorises ML algorithms based on their primary functions: time series networks for traffic prediction, federated learning for security and privacy, supervised learning for slice admission control, and reinforcement learning for resource allocation and management. It also highlights the complementary nature of these techniques, as demonstrated by hybrid models such as federated deep RL and LSTM-integrated distributed deep RL.Item Intelligent system for conformity assessment testing on the energy performance of household appliances—A South African perspective(2025) Molepo, IK; Markus, E; Abu-Mahfouz, Adnan MIThe potential of AI techniques to enhance conformity assessment processes has garnered attention within the sector. Traditional methods of conformity assessment suffer from a lack of critical analytics and susceptibility to subjective data analysis, human error, inefficiency, and poor data quality. Intelligent systems will improve the accuracy and consistency of assessments when evaluating the energy performance of household appliances. This paper presents some ML approaches to energy consumption anomaly detection applied to household appliances, with the potential to improve the efficiency, accuracy, and reliability of predicting energy consumption. The study proposes an intelligent system to enhance the accuracy, precision, and quality of appliance energy performance conformity assessment testing processes with minimal human intervention. Thus, providing a quality energy assessment, aiding in the efforts to reduce energy consumption in households.Item Developing a multi-modal biometric system for children using age-weighted feature matching approach(2025) Ntshangase, Cynthia S; Baruni, Kedimotse P; Holder, Ross P; De Kock, Antonie J; Khumalo, Lethukuthula P; Chossa, BhekiThe world is getting transformed by digital technologies to ensure balanced development. Likewise, the challenges introduced by the presence of these technologies are increasing. This includes the problem of the unavailability of digital biometric systems that can be used to verify the identity of children from birth to adulthood. This problem is caused by factors such as children’s biometric traits not being fully developed as those of adults and difficulty capturing high-quality biometrics from children using traditional systems. To solve this problem, this study proposes an age-weighted feature matching algorithm that fuses multiple biometric modalities. The system aims to improve the accuracy of identity verification for children. Experimental results show that the proposed multi-modal system, with the age-weighted algorithm, maintains higher recognition rates compared to individual biometrics. The Equal Error Rate (EER) for the multi-modal system without age-weighting is 14.07%, compared to 12.65% with age-weighting applied, indicating a higher successful recognition rate with fewer wrong matches or failures to match. In comparison, the EER for ear recognition is 57.56%, for fingerprint recognition is 32.17%, and for iris recognition is 37.76%, all of which are significantly higher than the combined multi-modal system, indicating lower recognition rates when singular biometric modalities are used. These findings emphasize the system’s potential to contribute meaningfully to applications such as birth registration, healthcare, and child protection—domains where robust and inclusive identity systems are urgently needed for children.Item Nature-based Solutions for Climate Adaptation(2024) Davies, Sarah J; Ntshotsho, PhumzaNature-based solutions include ecosystem-based adaptation to climate change. True ecosystem-based adaptation encompasses the following: Helps people adapt to climate change; Uses biodiversity and ecosystems as part of the solution and is part of a broader adaptation strategy.Item Physical oceanographic indicators for proactive management and early warning of harmful algal blooms(2025-03) Mtetandaba, A; Veitch, J; Smith, Marié EHarmful algal blooms (HABs) represent a significant threat to both the environment and the economy, particularly in coastal zones where aquaculture is a vital industry. These blooms, caused by the rapid growth of algae, can produce toxins that are detrimental to marine life, lead to hypoxic conditions (low oxygen levels), and contaminate seafood, posing risks to human health. The economic impact is substantial, as HABs can devastate fish populations, disrupt aquaculture operations, and result in costly closures of fisheries and tourism activities. Effective monitoring and prediction of HABs are therefore crucial for protecting marine ecosystems and ensuring the sustainability of aquaculture practices.Item Overview of the use of the ESA Sentinel-1 radar and -2 optical images for mapping and monitoring wetland biodiversity in South Africa(2025-02) Van Deventer, Heidi; Naidoo, L; Hansen, COverview of the use of the ESA Sentinel-1 radar and -2 optical images for mapping and monitoring wetland biodiversity in South AfricaItem Scoping community resilience concepts relevant to a South African definition for resource constrained environments(2018-08) Dlamini, Sifiso B; Herselman, Martha E; Marais, Mario AThe focus and aim of this paper is to indicate what aspects should be considered when developing a definition of community resilience for the South African context, with a special focus on using ICT in rural educational contexts. The aspects will be constituted by applying a scoping literature review ranging from 2014-2018 and examine the size and scale of research activities on the following concepts/search terms: resilience, social resilience, community resilience, community focused resilience, adaptive capacity and social capital. We investigated international electronic databases including ACM digital library, Scopus, IEEE Xplore, Harzings software and ScienceDirect, ISI Web of Science and Google Scholar. The search terms were using the topic of a paper, the keywords as well as the abstracts. Extracted data were analyzed in terms of resilience definitions, types of resilience, domains, and community resilience assessment. Results: A table with the most important concepts to consider were developed based on the scoping review where only 17 papers were found to be best suited to the purpose of the paper out of 675 publications initially identified. These produced an interesting list of aspects to consider with relating to the definition of community resilience which can be applied when developing this definition specifically for a South African context where ICT is applied in rural education. Considerable inter dependency between and among aspects of the concept community resilience also emerged from this analysis and a Wordle is provided to summarize the most essential aspects to consider at the end of the paperItem The benefits of replacing the traditional dimensional calibration system with the Laser Measurement Technology in the CSIR Wind Tunnel Facilities(2024-09) Ragimana, Phumudzo; Mabeko, Philimon K; Dikgale, Moyahabo SCalibrations in any aerodynamic metrology field require patience as it is time-consuming. The Council for Scientific and Industrial Research has wind tunnel facilities with a wind tunnel model support system of 6 degrees of freedom. These degrees of freedom are three translations and rotations, which must be calibrated per project-specific requirements and the facilities maintenance management schedule. This paper addresses the impact of introducing advanced laser measurement technology instead of traditional dimensional instrumentation calibration systems at the CSIR Wind Tunnel Facilities. Conventional calibration methods typically entail extensive manual processes and significant downtime, resulting in higher operational costs and slower instrumentation calibration and wind tunnel testing cycles. The advantage of implementing the laser tracker system is that it enhances the calibration accuracy in the CSIR Wind Tunnel facilities. Therefore, the laser tracker system addresses these challenges and offers an efficient alternative. By enhancing precision and speed, the new technology improves the wind tunnel facilities’ dimensional instrumentation maintenance processes and aerodynamic testing while delivering substantial cost and time savings. Manual and time-intensive traditional methods incur higher costs, slower calibration processes, and limited precision and human errors. Laser measurement technology simplifies these tasks, producing faster and more precise calibrations. Through several calibration activities and data collection, it is evident that this replacement enhances instrumentation calibration efficiency and decreases operational expenses, representing a valuable improvement for future aerospace research and development.Item The benefits of replacing the traditional dimensional calibration system with the Laser Measurement Technology in the CSIR Wind Tunnel Facilities(2024-09) Ragimana, Phumudzo; Mabeko, Philimon K; Dikgale, Moyahabo SCalibrations in any aerodynamic metrology field require patience as it is time-consuming. The Council for Scientific and Industrial Research has wind tunnel facilities with a wind tunnel model support system of 6 degrees of freedom. These degrees of freedom are three translations and rotations, which must be calibrated per project-specific requirements and the facilities maintenance management schedule. This paper addresses the impact of introducing advanced laser measurement technology instead of traditional dimensional instrumentation calibration systems at the CSIR Wind Tunnel Facilities. Conventional calibration methods typically entail extensive manual processes and significant downtime, resulting in higher operational costs and slower instrumentation calibration and wind tunnel testing cycles. The advantage of implementing the laser tracker system is that it enhances the calibration accuracy in the CSIR Wind Tunnel facilities. Therefore, the laser tracker system addresses these challenges and offers an efficient alternative. By enhancing precision and speed, the new technology improves the wind tunnel facilities’ dimensional instrumentation maintenance processes and aerodynamic testing while delivering substantial cost and time savings. Manual and time-intensive traditional methods incur higher costs, slower calibration processes, and limited precision and human errors. Laser measurement technology simplifies these tasks, producing faster and more precise calibrations. Through several calibration activities and data collection, it is evident that this replacement enhances instrumentation calibration efficiency and decreases operational expenses, representing a valuable improvement for future aerospace research and development.Item Digital decision support tools in EdTech testing and evaluation for developing countries(2024-12) Mnyandu, Wandile T; Terzoli, A; Kobo, Hlabishi IInformation and communication technologies (ICTs) are essential the integration of educational technologies (EdTech) into modern classrooms, but many developing countries lack the IT Infrastructure necessary to support these tools. This paper aims to facilitate the use of these tools by reviewing ICT devices commonly deployed in public schools, focusing on the features and functionalities that can be effectively tested using DDSTs.Item Climate Change Risk and Vulnerability Assessments and Adaptation strategies for South African Ports(2024-04) Van Ballegooyen, R; Siyakatshana, Njabulo; Engelbrecht, F; Daniels, T; Beakam, V; Rossouw, M; Mateyisi, Mohau J; Troch, Christophe NA; Zwane, Nosipho; Smith, G; Mbatha, NSouth Africa's ports play a major role in the country's economy. Should climate-related infrastructure damage or operational disruption occur, this could have severe "knock-on" effects to the wider economy. Consequently, Transnet, since 2010, has been in the process of developing an integrated Climate Change Strategy across its operations, commencing with a preliminary Risk and Vulnerability assessment for all Transnet Operating divisions in 2010, followed in 2014 by a first-level risk and vulnerability assessment for the Transnet National Port Authority to assess potential drivers of climate-change risk to port infrastructure, operational support systems, activities related to safety and long-term sustainability of the eight ports they presently manage. Only once the climate-related changes in these key environmental drivers are adequately understood and characterised at a local scale, is it possible to assess climate change risks in each port with sufficient rigour to ensure that any proposed adaptation measures are both appropriate and economically justified. Presented here is initial progress with a TNPA-funded, multi-year, collaborative effort by a consortium of scientist and engineers from range of local institutions and consultancies in making such a detailed Climate Change Risk and Vulnerability Assessment informing the development of Adaptation strategies for South African Ports.Item Enhancing dynamic spectrum access in wireless mesh networks: Integrated sensing and channel assignment(2024-12) Zlobinsky, N; Lysko, Albert AThis study advances Channel Assignment (CA) strategies in wireless mesh networks that integrate Wi-Fi with Dynamic Spectrum Access (DSA). We propose critical modifications to the IEEE 802.11 standard to enhance spectrum sensing data reporting, which is crucial for efficient network resource management. Specifically, we introduce two novel measurement types– avgSINR and SINRCI– and detail their integration into network protocols to optimise CA. Furthermore, we outline robust procedures for the dynamic distribution and adjustment of CAs, ensuring rapid network adaptation. Our findings demonstrate that the network can rapidly converge on a new CA within 10 minutes, even in extensive networks comprising 100 nodes, thereby significantly enhancing operational efficiency and spectrum utilisation.Item Securing individual’s identity using ZKP: Proposed system architecture(2024-11) Ntshangase, Cynthia S; Baruni, Kedimotse P; Myaka, Zanele S; Mahlasela, Oyena NDigital identity with biometrics is the future of how individuals will be identified in this digital era. Although there are current solutions that provide digital identity, the research reveals that traditional solutions lack security for biometric data, making them vulnerable to biometric data breaches, and data lost with difficulties to recover. This study discusses the importance of protecting our identities, as identity theft is increasing rapidly due to the rise of identity verification online. A comprehensive review on protecting identity using biometrics and Zero Knowledge Proofs has been conducted. A novel biometric system architecture based on Zero Knowledge Proofs has been proposed, ensuring that verification of an identity is performed without exposing biometric data. This architecture also includes a reference data engine for updating biometrics data in case of a breach, to improve security and overcome challenges faced by traditional biometric systems. The proposed architecture can be used to secure identities while preserving privacy.Item A perspective on the integration of ABB industrial robots into Open AI Gym using vision-based reinforcement learning(2024-12) Dikole, Realeboga G; Faniso-Mnyaka, Zimbini; Sekopa, Teboho LReinforcement learning-based robotic manipulation has risen to prominence recently due to the rise of frameworks such as Open AI Gym. Although there has been much success in reinforcement learning applied to manipulators such as the Kuka iiwa, Franka Emika Panda, UR5 robots, there has been little to no exploration of the application of reinforcement learning on ABB robotic manipulators. This paper presents a perspective on integrating vision-based reinforcement learning and ABB robots with Open AI Gym environments. We focus mainly on the pick-and-place and push environments using proximal policy optimisation due to its simplicity and ease of memory. Our results affirm the possibility of vision-based reinforcement learning with ABBrobots with the best training performance yielding a success rate of close to sixty-five percent.Item Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems(2025-01) May, N; May, N; Bokoro, P; May, Siyasanga I; Mkasi, Hlaluku WThis research investigates the effectiveness of machine learning and deep learning models in forecasting voltage swell peak amplitudes within grid-connected photovoltaic (PV) systems, aiming to enhance power quality management. A 24month dataset (January 2022 - December 2023) encompassing power and weather data from a 3.3 kWp PV system at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa, was utilized. Hourly averaged data between 5 am and 6 pm, capturing PV system and weather measurements, was analysed. Five models – Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM) – were trained and evaluated using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The Random Forest model demonstrated superior predictive accuracy, closely aligning with actual peak voltages and achieving the lowest MSE (0.01V²) and RMSE (0.02V). This study highlights the potential of machine learning, particularly Random Forest, in accurately predicting voltage swell peak amplitudes, contributing to improved power quality management in grid-connected PV systems.Item Investigating the use of the adiabatic calorimetry to characterize concrete road pavement mixes(2024) Moasa, Boipelo M; Mokoena, Refiloe; Pitso, Reatile IDetermining in-situ concrete properties can be a challenge due to limitations on current test methods where laboratory conditions and in-situ conditions are not the same. The heat released from cementitious materials during the exothermic hydration process can be used to characterize important properties of different mix designs for concrete pavements. In this investigation, an adiabatic calorimeter for concrete mixes was used to measure the real time heat of hydration of concrete mixes for road pavements. The adiabatic calorimeter captures the true temperature of concrete specimens with minimal heat exchange between the concrete and surroundings to determine real-time concrete heat properties. This study investigates three different concrete pavement mix designs that incorporate CEM II 32.5 R, CEM II 42.5 N and CEM I 52.5 N cements respectively. The heat of hydration and the correlation with concrete properties such as 28-day flexural strengths, 7-day, and 28-day compressive strengths, initial and final setting times are discussed in this paper. The results indicate a high correlation (0.91Item Drawbacks of pothole filling programs as a preventitive maintenance measurea: Study based on non-intrusive pavement defects investigation(2024-09) Mulol, Christian ARecent studies have alerted the South Africa Road Sector and its various stakeholders to the alarming deterioration rate of the road network condition. This is especially the case with the secondary and the tertiary road network where the backlog of maintenance is in the order of hundreds of billions of Rands. Amongst other reasons such as the lack of technical capacity within the secondary and tertiary road authorities, this backlog of maintenance has often been named as one of the major contributors to the poor rating of the lower ranking roads in South Africa. Roads within the secondary and tertiary sectors are either at risk of failure, being unable to cope with the normal demand, subjecting the public to severe inconvenience, or unfit for purpose, having already failed or being on the verge of failure, exposing the public to health and safety hazards. To ensure all the safety hazards on the road network are removed with immediate effect, preventive maintenance through the filling of potholes can be recommended. However, this approach was never intended to be the substitute for other maintenance or rehabilitation measures. This paper shows that though the preventive maintenance through the filling of potholes has a merit from a safety point of view, this method does not necessarily address the root-cause that led to the initiation and the generation of such potholes. Using various roads as case studies and through a comparison of their TMH 9 visual assessment strip map to the FWD deflection parameters mapping, this paper recommends that the reactive maintenance should not only follow the rigorous requirements of pothole repairs that extend into the base and underlying layers but should also be taken as a holding measure guaranteeing the safety of motorists until the appropriate rehabilitation or maintenance measure is implemented without further delay.Item Future climatic considerations for performance based criteria of South African concrete pavements(2024-08) Mokoena, Refiloe; Mturi, G; Manganyi, ObeyThis study is focused on identifying key climatic variables and considerations for the design of concrete pavements to cater for changing climatic conditions. To mitigate the effects of thermal distresses, such as cracking and curling of concrete pavements, pavement engineers consider factors such as reinforcement detailing, joint location and spacing. Typically, pavement design approaches consider historical temperature records in estimating the daily variation in air temperature over the service life of concrete pavements. Current practice does not incorporate future climate projections and urban heat island (UHI) effects that can affect both the safety and serviceability of concrete pavement infrastructure. Therefore, this article discusses the need to investigate concrete pavement deterioration in relation to projected global warming effects for South Africa. To demonstrate this, a discussion of micro- and meso-structural effects is presented. Furthermore, a simulation is included that uses future climate projections to estimate the change in maximum diurnal temperature range (DTR) and annual rainfall for the duration of a Jointed Plain Concrete Pavement's service life in Pretoria, South Africa. To ensure the construction of climate resilient concrete pavements that can withstand expected service conditions of projected environmental conditions, this future-based climatic input will need incorporation into performance-based tests and specifications.