Research Articles (Electrical, Electronic and Computer Engineering)

Permanent URI for this collectionhttp://hdl.handle.net/2263/1693

Browse

Now showing 1 - 20 of 1021

FPEVO : fused point-edge visual odometry for low-structured and low-textured scenes
Brown, Dylan; Grobler, Hans; De Villiers, Johan Pieter (Elsevier, 2025-11)
Visual odometry is an essential component of vision-based robotic navigation systems. A primary limitation of existing visual odometry solutions is their inability to achieve satisfactory performance in both high- and low-textured regions. In this paper, a robust RGB-D visual odometry method is proposed that fuses point and edge features. By combining the descriptiveness of feature points with the structure provided by edge data, a method that is robust to low-textured scenes is developed. Edge features are first detected and grouped based on the Gestalt principles of continuity and proximity. Edge groups are then associated between the current and previous frames using point features in the vicinity of the edges. Pose estimation is thereafter performed by first matching points between associated edge groups, filtering these points based on structural constraints imposed by the edges, and estimating the motion of the agent. Compared to state-of-the-art alternatives, such as REVO, MSC-VO, DROID-VO and SplaTAM on the TUM RGB-D, ICL-NUIM and Tartan-Air datasets, the resulting method reduces the root mean square absolute trajectory error, and translational and rotational relative pose errors by up to 58%, 75%, and 82%, respectively. This indicates that our method is not only more accurate than current approaches, but also more consistent, especially in low-structured and low-textured environments.
Integrating computational modelling into the ecosystem of cochlear implantation : advancing access to diagnostics, decision-making, and post-implantation outcomes on a global scale
Hanekom, Tania (MDPI, 2025-11-08)
Disabling hearing loss affects more than 5% of the global population, with numbers expected to double by 2050. The burden is especially high in low- and middle-income countries, where access to cochlear implant (CI) technology and the required follow-up care is limited. While CIs are a proven treatment for certain types of hearing loss, their adoption in these countries is hindered by high costs, the need for specialised rehabilitation, and the financial and time commitment required for long-term device maintenance. Although remote programming has improved accessibility to standard care, specialised interventions for complications remain restricted mainly to areas with clinical centres. Computational modelling offers a promising solution to this access-to-care dilemma. The models may be used to simulate complications, such as non-auditory stimulation (NAS), to investigate and plan personalised interventions, and ultimately predict device parameters, without requiring the recipient’s physical presence. Both phenomenological and biophysical models have already demonstrated useful application in CIs: the former streamlines clinical workflows and aims to establish consistency in device fitting, and the latter provides insights into patient-specific auditory biophysiology. Despite decades of research, clinical translation of biophysical models has been limited by data constraints, parameter uncertainty, and validation challenges. In this perspective piece, it is argued that biophysical models have now reached sufficient maturity to be integrated into routine CI care. Apart from the advantages that this approach will bring to the overall advancement of person-centred CI care, it is envisioned to improve accessibility, personalisation, and long-term outcomes for CI recipients in low- and middle-income countries.
Two-stage optimization of appliance scheduling and BESS capacity with comfort level
Ren, Zhiling; Chen, Xi (Dong, Yun, 2026-06)
Residential photovoltaic (PV)–battery energy storage system (BESS) planning often neglects impact of comfort level. We propose a two-stage framework that jointly selects BESS capacity and appliance start-up schedules while accounting for appliance-use comfort, thermal comfort, and BESS degradation. Annual days are clustered by k-means to select a set of typical days. Stage 1 performs a discrete capacity scan (coarse grid plus local refinement). For each candidate, Stage 2 solves a typical day mixed-integer linear programming problem with a linear fractional cost–comfort objective via Dinkelbach’s method, maps the resulting binary schedules to all natural days by cluster membership, and then solves an annual rolling linear programming problem for dispatch and degradation-inclusive evaluation. In the baseline case, the selected capacity is 10.75 kWh with an annual total cost of 2623.2 CNY and comfort indices of 0.793 and 0.788; degradation cost is 404.3 CNY. Without the comfort term, the preferred capacity increases to 16.5 kWh and the annual total cost decreases to 2265.9 CNY, while degradation cost rises to 555.1 CNY. Sensitivity analyses show that outcomes vary with comfort settings and time-of-use prices. Overall, the framework quantifies annual cost-comfort-degradation trade-offs and recommends a feasible BESS size and appliance schedules. HIGHLIGHTS • Two-stage framework jointly optimizes BESS capacity and appliance schedules with comfort. • k-means typical days enable tractable annual MILP and rolling LP evaluation. • Baseline selects 10.75 kWh, 2623.2 CNY, with comfort indices 0.793 and 0.788. • Removing comfort changes preferred BESS capacity and appliance start-up schedules. • Sensitivity analyses quantify impacts of comfort weights, bounds, price.
A carbon subsidy framework for coordinated low-carbon operation in industrial park with multiple users
Ren, Zhiling; Chen, Xi; Zhao, Boya; Dong, Yun (Elsevier, 2026-05)
As the global impacts of carbon emissions intensify, power markets are shifting from profit-driven models to frameworks that integrate economic performance and carbon mitigation. Current low-carbon approaches commonly exhibit insufficient user response, elevated abatement costs, complex implementation, and limited flexibility. To bridge this gap, we propose a carbon subsidy framework for industrial parks. In this framework, the Industrial Park Operator (IPO) announces real-time shared electricity, carbon, and subsidy prices, while users optimize load scheduling and market participation. The interaction is formulated as a Stackelberg game and solved using a Differential Evolution-Mixed-Integer Quadratic Programming (DE-MIQP) approach. Case studies demonstrate that, compared with a baseline without subsidies, the IPO provides a total subsidy of 1,295.57 CNY, the proposed framework reduces industrial-park CO2 emissions by 40.13 %, and the IPO’s profit decreases by 217.39 CNY, while the users’ aggregate profit increases by 608.3 CNY. Finally, two sensitivity analyses are conducted: one investigates parameter sensitivity, and the other is a ten-user case study that exhibits emission-reduction behavior similar to that of the four-user case. These findings confirm that the carbon subsidy framework effectively incentivizes user participation, reduces emissions, and offers a practical pathway for coordinated low-carbon operation in multi-user industrial parks. HIGHLIGHTS • Carbon subsidy framework for coordinated low-carbon industrial park operation. • Stackelberg game with DE-MIQP preserves user privacy while finding equilibrium. • Subsidy cuts park CO2 emissions by 40.13 % and SPP output by 47.01 % versus baseline. • Profit shifts: IPO -217.39 CNY, users +608.3 CNY, subsidies total 1,295.57 CNY. • Sensitivity and ten-user studies confirm robustness and scalability of the scheme.
A bi-level optimization framework for virtual power plants integrating electric vehicles and demand response
Ren, Zhiling; Li, Shaoming; Guo, Jia; Lin, Dong; Dong, Yun (Elsevier, 2025-12)
The increasing penetration of wind and photovoltaic (PV) generation introduces significant uncertainty and volatility to power systems. To address these challenges, this study proposes a bi-level optimization framework for virtual power plants (VPPs) that integrates electric vehicles (EVs) and demand response (DR) to enhance renewable energy utilization, reduce carbon emissions, and coordinate the economic interests between the VPP operator (OPE) and the aggregator (AGG). The upper level maximizes the OPE’s revenue through dynamic electricity pricing, while the lower level minimizes the AGG’s cost via adaptive load scheduling. Simulation results show that, compared to a baseline case without EV and DR coordination, the proposed framework reduces peak demand by 3.02%, lowers total carbon emissions by 10.13%, and decreases renewable energy curtailment by 37.5% for wind and 42.85% for PV. To further validate robustness, the model was tested under diverse weather conditions over a one-week period, achieving even greater reductions in curtailment: 51.07% for wind and 51.38% for PV. This framework provides a scalable solution for high renewable integration, enabling both economic and environmental benefits.
A critical assessment of cable rating methods under soil drying out conditions
Khumalo, Ntombifuthi Queeneth; Naidoo, Raj M.; Mbungu, Nsilulu T.; Bansal, Ramesh C. (Wiley, 2025-10)
The design of underground cable systems must account for the risk of soil drying out due to heat dissipation, which can degrade cable performance and lead to environmental concerns. This study investigates a cost-effective cable rating methodology tailored to South African conditions, where native soils are used instead of engineered backfill. Using the IEC 60287 standard, an Excel-based calculation tool is developed to assess the effects of key installation parameters, including soil thermal resistivity, ambient soil temperature and cable laying depth. Soil samples from Sandton, South Africa, revealed thermal resistivity ranging from 0.596 K·m/W, at 14.5% moisture, to 3.72 K·m/W, at 0% moisture, resulting in current ratings from 518.34 A to 224.21 A. Worst-case conditions—high resistivity, increased depth, 1150 mm and elevated soil temperature, 28°C—reduced ampacity by over 45%. The findings underscore the need to incorporate site-specific soil data and worst-case assumptions into cable rating designs to prevent thermal degradation. The developed method offers a practical, locally optimised alternative for utilities in semiarid regions.
A load-balancing enhancement to schedule-aware bundle routing
Kamps, Jason Jack; Paluncic, Filip; Maharaj, Bodhaswar Tikanath Jugpershad (Wiley, 2025-03)
Delay- and disruption-tolerant networking (DTN) enables communication in networks afflicted by long propagation delays and sporadic connectivity. DTN routing techniques such as schedule-aware bundle routing (SABR) exist to route data bundles in de-terministic networks, such as those found in deep-space environments, where node contacts are predictable. This article begins with an overview of DTN architecture and SABR. SABR's method of final route selection (forwarding rules) is closely examined. The article then addresses a limitation of SABR whereby the algorithm may overlook parallel channels, leading to network congestion. To mitigate this, an enhancement is proposed. This enhancement aims to optimize data bundle distribution across candidate routes in networks with parallel channels, thus alleviating congestion and enhancing overall network performance. This is achieved with simple modifications to SABR's forwarding rules to avoid the concentration of data bundles on a minority of node contacts. The enhancement is demonstrated through simulations in a reference scenario implemented in DtnSim.
Earth grid : toward a low-carbon energy infrastructure
Kumar, Abhishek; He , Xiangning; Deng, Yan; Sah, Bikash; Singh, Arvind R.; Kumar, Praveen; Bansal, Ramesh C.; Kirtley, James L. (Cell Press, 2025-11-21)
With fossil fuels as a major global energy source and their associated carbon emissions impacting the climate, there is an urgency in transitioning to carbon-free energy sources (CFESs) such as nuclear and renewables (solar, hydroelectric, and wind). The need for an updated electric power transmission and distribution system arises due to the variable loads and geographic diversity of renewable energy sources. This perspective explores the idea of the “Earth Grid,” which proposes an intercontinental electric grid facilitated by three technological advancements: enhanced information and communication technology (ICT) applications in the electric grid, development of inter-country grids for power-sharing, and the application of artificial intelligence (AI) for efficient operation and maintenance. Further, the article discusses the need for a collaborative effort on the global stage to transform the existing electricity generation, transmission, and distribution sector to create a globally interconnected carbon-neutral Earth Grid infrastructure for accelerated carbon-free energy access for all.
Multiauthority KP-ABE access model with elliptic curve cryptography
Ferrer-Rojas, Agustin; Maharaj, Bodhaswar Tikanath Jugpershad (South African Institute of Electrical Engineers, 2025-06)
The rapid and expansive integration of Internet of Things (IoT) environments across various industrial sectors has led to an unprecedented surge in data generation and management. This exponential growth in data underscores the critical necessity for robust data security methodologies that can effectively safeguard the confidentiality and integrity of information without imposing undue computational burdens. In response to this challenge, numerous studies have sought to leverage Attribute-Based Encryption (ABE) as a means to enable fine-grained access control. Among the ABE variants, Ciphertext Policy ABE (CP-ABE) and bilinear pairings have emerged as popular choices to construct security schemes that strike a balance between robust protection and computational efficiency. Despite the advancements achieved through CP-ABE and bilinear pairings, a prevalent concern arises in the utilization of Linear Secret Sharing Scheme (LSSS) access policies. LSSS policies, while providing a flexible and expressive way to define access controls, can significantly impact the execution time of encryption methods. This study recognizes the importance of addressing this challenge and explores the potential of employing a Key Policy Attribute-Based Encryption (KP-ABE) approach. The primary objective is to mitigate the computational overhead associated with encryption methods, thereby enhancing the efficiency of data security measures within IoT environments. Furthermore, this research delves into the incorporation of Elliptic Curve Cryptography (ECC) to generate cryptographic keys. ECC, known for its strong security properties and computational efficiency, is considered a promising approach to bolster data security while concurrently minimizing computational overhead. By integrating KP-ABE with ECC, this study aims to offer a comprehensive solution that ensures robust security measures within the intricate landscape of IoT environments. Through detailed analysis and empirical investigation, the research endeavors to contribute valuable insights to the ongoing discourse on securing IoT data in a manner that aligns with the dual imperatives of security and computational efficiency.
Point process models for predicting the spatial distribution of rhino poaching activity in the Kruger National Park
Kirkland, Lisa-Ann; Fabris-Rotelli, Inger Nicolette; De Villiers, Johan Pieter (South African Statistical Association, 2025-09)
Rhino poaching in South Africa continues to threaten the existence of African rhino species. Since poachers often attack wildlife parks frequently, predictive models are essential for exploiting the availability of data to gain information about the poachers. Although a number of statistical methods have been applied to poaching prediction, they either do not take the spatial variation of observations into account, require additional observational data, depend on known priors, or result in models that are overfitted and challenging to interpret. This paper proposes the use of point process models to predict the spatial distribution of poaching activity within a wildlife park. Descriptive statistics of poaching spatial point patterns have been considered, as well as univariate non-parametric kernel density estimation. However, the focus of this work is on fitting multivariate parametric point process models, using a number of environmental factors. Since real-world poaching data could not be obtained for this work, due to the sensitivity of the data, a simulation study is performed, where numerous point patterns are generated from the same underlying point process. The method can be used when no data is available, and is based on environmental preferences of poachers, which can be obtained through expert knowledge, literature reviews, or by making intelligent assumptions. The results indicate that the point process models are able to predict the initial probabilities well, for most data generating processes. Point process models thus appear to be a promising method for predicting the spatial distribution of poaching activity.
Situational awareness indices of solar PV power generation under temporal weather conditions for near real-time planning and operation
Walters, Michael; Venayagamoorthy, Ganesh Kumar (Elsevier, 2025-12)
Solar photovoltaic (PV) plant development and utilization is increasing worldwide but remains intrinsically challenged by its large dependence on highly variable weather conditions and operating states. This paper presents a framework to leverage three new situational awareness indices (SAIs), namely: weather condition index (WCI) to gauge operational performance based on environmental states, operational complexity index (OCI) to indicate the severity of power generation reductions, and photovoltaic generation index (PVGI) to provide a final determination of the impact on power generation and to bolster situational awareness in planning and operational contexts for solar PV plants. This is accomplished by exploiting the effects of weather conditions, operating states, and solar PV power generation performance in high spatial-temporal resolution contexts residing in solar PV power generation data with independent fuzzy inference systems (FISs) for each index. SAIs provide additional operational insights to evaluate solar PV plant performance over both short-term (minute(s)) and long-term (24 h) time intervals in a variety of areas, including weather condition classification studies, energy dispatch controllers, and power system voltage and frequency stability assurance. The proposed SAI framework is developed, demonstrated, and evaluated for a 1MWp solar plant located in Clemson, South Carolina, USA.
The Internet of Battle Things : a survey on communication challenges and recent solutions
Kufakunesu, Rachel; Myburgh, Hermanus Carel; De Freitas, Allan (Springer, 2025-01)
The use of Internet of Things (IoT) technology in military settings has introduced the notion of “Internet of Battle Things” (IoBT), transforming modern warfare by interconnecting various equipment and systems essential for battlefield operations. This connectivity facilitates real-time communication, data sharing, and collaboration among military assets, enhancing situational awareness, decision-making processes, and overall operational effectiveness. The domain for IoBT encompasses a broad range of military assets, from drones and ground vehicles to soldier-worn wearables, sensors, and munitions. These assets are capable of collecting and transmitting critical information from the battlefield, including location data, status updates, environmental conditions, and the movements of adversaries. IoBT networks depend on robust communication networks, secure data transmission protocols, advanced data analytics for processing vast datasets, and seamless integration with command-and-control infrastructures. However, IoBT devices and systems function in dynamic and challenging battlefield conditions which present unique communication challenges. This study aims to review research efforts that provide current state-of-the-art solutions, their limitations, and emerging technologies. We classify these challenges into interoperability, power and energy management, security, and network resilience, while also discussing future research directions to improve communication in IoBT networks.
Reinforcement learning based automatic tuning of PID controllers in multivariable grinding mill circuits
Van Niekerk, Jonathan Anson; Le Roux, Johan Derik; Craig, Ian Keith (Elsevier, 2025-12)
Process controllers are extensively utilised in industry and necessitate precise tuning to ensure optimal performance. While tuning controllers through the basic trial-and-error method is possible, this approach typically leads to suboptimal results unless performed by an expert. This study investigates the use of reinforcement learning (RL) for the automatic tuning of proportional–integral–derivative (PID) controllers that control a grinding mill circuit represented by a multivariable nonlinear plant model which was verified using industrial data. By employing the proximal policy optimisation (PPO) algorithm, the RL agent adjusts the controller parameters to enhance closed-loop performance. The problem is formulated to maximise a reward function specifically designed to achieve the desired controller performance. Agent actions are analytically constrained to minimise the risk of closed-loop instability and unsafe behaviours during training. The simulation results indicate that the automatically tuned controller outperforms the manually tuned controller in setpoint tracking. The proposed approach presents a promising solution for real-time controller tuning in industrial processes, potentially increasing productivity and product quality while reducing the need for manual intervention. This research contributes to the field by establishing a robust framework for applying RL in process control, designing effective reward functions, constraining the agent to a safe operational space, and demonstrating its potential to address the challenges associated with PID controller tuning in grinding mill circuits.
Locomotive wheel-slip control with slip ratio reference adaptation using model-based estimation
Van de Merwe, Charl Viljoen; Le Roux, Johan Derik; Limebeer, David J.N. (Elsevier, 2025-12)
This paper investigates the wheel-slip control of locomotive traction systems in the presence of uncertain wheel-rail rolling contact conditions. A linear estimator is used to produce estimates of the wheels’ slip ratios and adhesion coefficients. These estimates are used as part of a slip ratio reference adaptation scheme that provides a reference to an adaptive PI controller. The control architecture is intentionally designed to be suitable for practical deployment in industrial settings, where simplicity and reliability are essential. A detailed pitch-plane simulation model is used to validate the controller performance. The results indicate that the estimator-controller combination can prevent unstable slip over a wide range of adhesion conditions, thereby preventing damage to the wheels and rail while ensuring maximum adhesion.
Model-plant mismatch diagnosis using plant model ratios for a grinding mill circuit under model predictive control
Mittermaier, Heinz Karl; Le Roux, Johan Derik; Craig, Ian Keith (Elsevier, 2025-07)
Model-based controllers often extend improved performance to mineral processing plants by leveraging predictive models to account for system dynamics, handling constraints, adapting to changing conditions, and optimizing control inputs. Inaccurate models will cause a deterioration of controller performance, which is often the case for grinding mill circuits. The plant model ratio was developed to diagnose parametric model plant mismatches for first-order plus time delay models. Using a simulation study, the plant model ratio is applied to test the feasibility of using the plant model ratio on a grinding mill circuit. By applying different scenarios of mismatch, some limitations of the plant model ratio are identified and discussed in light of a grinding mill circuit model that is used in model-based controllers. The plant model ratio is capable of identifying parametric model plant mismatches for the model of a grinding mill circuit, specifically changes in the direction of responses. This may occur in cases where disturbances push a grinding mill to operate to the right of the peak of a grind curve. HIGHLIGHTS • Inaccurate process models deteriorate model-based controller performance. • Plant model ratio (PMR) can identify model-plant mismatch (MPM) for MIMO systems. • A simulation study shows how PMR can identify MPM for a grinding mill circuit. • PMR is effective to identify changes in the response direction of process variables.
Dual phase-shift symmetrical SVM strategy optimized by dual modulation indices for single-stage isolated DC–AC matrix converter
Ye, Guangcheng; Qin, Hao; Kumar, Abhishek; Bansal, Ramesh C.; Gryazina, Elena; Deng, Yan (Institute of Electrical and Electronics Engineers, 2026-01)
The single-stage, isolated three-phase dc–ac converter presents significant advantages for applications involving low dc voltages. By eliminating the need for large capacitors, this design facilitates a more compact physical layout. However, the complexity inherent in modulation strategies frequently gives rise to substantial three-phase current harmonics, which adversely affect system performance. Despite numerous efforts documented in the literature to enhance various performance metrics, the effectiveness of these solutions has often been found to be limited. This article presents a new modulation strategy designed to enhance the overall performance of high-frequency link matrix converters. We utilize frequency domain analysis through Fourier transform for our modeling approach. Our strategy optimizes performance by employing a combination of dual shift phase angles and dual modulation indices. Finally, the validation of this strategy was conducted on a three-phase 800 W prototype. Experimental outcomes demonstrated a Total Harmonic Distortion below 2% using a single DSP controller, corroborating theoretical model which predicted a reduction in current stress to 84% of its initial level, and the peak value of low-frequency fluctuation is reduced to 72–78%.
Non-linear control of a fuel gas blending benchmark problem with added consumer dynamics
Sibiya, M.D.; Wiid, Andries Johannes; Le Roux, Johan Derik; Craig, Ian Keith (Elsevier, 2025-10)
This paper contributes to existing literature on fuel gas control by providing a feasible control solution with improved economic performance for an existing fuel gas control benchmark problem. Improved economic performance is achieved by implementing a non-linear model predictive controller (NMPC) that uses state estimates provided by a moving horizon estimator (MHE) and extended Kalman filter (EKF) for the fuel gas composition and flame speed index (FSI) to provide continuous inputs for the controller. Furthermore, the original fuel gas benchmark model is expanded to include consumer dynamics affecting fuel gas demand due to changes in the fuel gas heating value, making the model more representative of real industrial plants. The behaviour of an NMPC that neglects consumer dynamics (NMPC1) was compared against an NMPC that includes consumer dynamics (NMPC2). The aim of the benchmark problem is to reduce the time-weighted average cost of fuel gas for three 46-hour cases, accounting for purchase costs and penalties for fuel gas specification violations. An optimal cost for each case is determined assuming ideal conditions and perfect control. The benchmark controller is a conventional multi-loop feedforward/feedback system and has an average cost for the three cases which is 38.5% higher than the optimal cost. The NMPC1 controller has an average cost which is 33.9% higher than the optimal cost and better than the benchmark controller. A new benchmark scenario was developed which includes the consumer dynamics. For the new scenario, NMPC1 could not find a feasible solution, resulting in oscillations and specification violations. The oscillations would result in site-wide instabilities for all equipment using fuel gas. NMPC2 was able to keep the process stable during these scenarios and maintain all specifications. This shows the necessity to include consumer dynamics for effective fuel gas blending control. HIGHLIGHTS • Show improved economic performance for an existing fuel gas control benchmark problem. • Provide continuous estimates of fuel gas composition and flame speed index to an NMPC. • Consumer dynamics are included, making the model more representative of industry. • An NMPC that neglects consumer dynamics is compared against one that does not. • Show the necessity to include consumer dynamics for effective fuel gas blending control.
A review of smart crop technologies for resource constrained environments : leveraging multimodal data fusion, edge-to-cloud computing, and IoT virtualization
Olatinwo, Damilola D.; Myburgh, Hermanus Carel; De Freitas, Allan; Abu-Mahfouz, Adnan Mohammed (MDPI, 2025-10-09)
Smart crop technologies offer promising solutions for enhancing agricultural productivity and sustainability, particularly in the face of global challenges such as resource scarcity and climate variability. However, their deployment in infrastructure-limited regions, especially across Africa, faces persistent barriers, including unreliable power supply, intermittent internet connectivity, and limited access to technical expertise. This study presents a PRISMA-guided systematic review of literature published between 2015 and 2025, sourced from the Scopus database including indexed content from ScienceDirect and IEEE Xplore. It focuses on key technological components including multimodal sensing, data fusion, IoT resource management, edge-cloud integration, and adaptive network design. The analysis of these references reveals a clear trend of increasing research volume and a major shift in focus from foundational unimodal sensing and cloud computing to more complex solutions involving machine learning post-2019. This review identifies critical gaps in existing research, particularly the lack of integrated frameworks for effective multimodal sensing, data fusion, and real-time decision support in low-resource agricultural contexts. To address this, we categorize multimodal sensing approaches and then provide a structured taxonomy of multimodal data fusion approaches for real-time monitoring and decision support. The review also evaluates the role of IoT virtualization as a pathway to scalable, adaptive sensing systems, and analyzes strategies for overcoming infrastructure constraints. This study contributes a comprehensive overview of smart crop technologies suited to infrastructure-limited agricultural contexts and offers strategic recommendations for deploying resilient smart agriculture solutions under connectivity and power constraints. These findings provide actionable insights for researchers, technologists, and policymakers aiming to develop sustainable and context-aware agricultural innovations in underserved regions.
A wideband circularly polarised magneto-electric dipole antenna array with a series sequential phase feed network
Coetzer, Elmien; Joubert, Johan; Odendaal, Johann Wilhelm (Wiley, 2025-01)
A printed circularly polarised antenna array is presented that utilizes the inherent good bandwidth and stable gain of magneto-electric dipoles in combination with the wideband benefits of a sequential rotation feed technique. The proposed antenna has a simple geometry using two substrates and does not require any additional cavity or parasitic elements. The designed and simulated antenna has an impedance bandwidth of more than 75%, a 3 dB axial ratio bandwidth of 67% and a peak gain of 12.4 dBic, with less than 3 dB gain variation across the entire axial ratio bandwidth. The antenna provides a good combination of simple and compact geometry, wide bandwidth, good gain and stable radiation patterns when compared to previously published research. Simulated as well as measured results are presented for a protype antenna array.
Hybrid intelligent optimisation for onshore wind farm forecasting
Gwabavu, Mandisi; Bansal, Ramesh C.; Bryce, Andrew (Springer, 2025-09)
Accurate wind power forecasting is crucial for the dependable functioning and strategising of contemporary power systems, especially as the global integration of renewable energy escalates. This study introduces an innovative hybrid intelligent forecasting model that amalgamates Long Short-Term Memory (LSTM) neural networks with Complementary Ensemble Empirical Mode Decomposition (CEEMD) and a hybrid optimisation strategy that incorporates Ant Colony Optimisation (ACO), Genetic Algorithm (GA), and Particle Swarm Optimisation (PSO). The model was developed and evaluated utilising empirical data from a 138 MW wind farm consisting of 46 turbines, based on operational data from 2019. The proposed CEEMD-LSTM-ACO-GA-PSO model adeptly tackles the nonlinearity and intermittency of wind speed data through the decomposition of intricate signals, the enhancement of temporal learning, and the optimisation of model hyperparameters. The evaluation results indicated a substantial enhancement in forecasting precision relative to baseline models. The hybrid model attained a Root Mean Square Error (RMSE) of 0.142 and a Mean Absolute Percentage Error (MAPE) of 3.8% for 24-h forecasts, representing an enhancement of more than 35% compared to traditional LSTM models. It also exhibited strong performance over extended forecasting periods of up to 168 h. This study validates the effectiveness of a hybrid intelligent model in improving wind power forecasting while emphasising the limitations associated with computational complexity, sensitivity, feature importance and generalisation. Future research should incorporate uncertainty quantification, simplify models for real-time deployment, and adopt transformer-based architectures. The results endorse the application of intelligent optimisation in enhancing the reliability and sustainability of energy system operations.

Browse

Recent Submissions