Research Articles (Electrical, Electronic and Computer Engineering)

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

Browse

Now showing 1 - 20 of 1059

Full-wave synthesis technique for the design of an optimized 0.5-18 GHz 90° hybrid coupler
Du Toit, Johannes Bartholomeus; Joubert, Johan; Odendaal, Johann Wilhelm (Wiley, 2026-04-09)
An optimization technique that can be used to design high performance, wideband 90° hybrid couplers is described. Optimization is performed not by using the traditional theoretical coupling factors, but rather by directly synthesizing the geometric dimensions of the hybrid in full-wave simulations. Simulated results thus include all secondary, nonideal transmission line and implementation effects, and can be optimized for the required equiripple results. The full-wave synthesis technique is explained in detail, and simulated results of a 2–18 GHz design with 0.5 dB magnitude imbalance improvement over any previous results are shown. It is also used to implement a unique 0.5–18 GHz 3 dB, 90° tandem hybrid, of which measurements are presented showing near optimally minimized magnitude imbalance of 1.47 dB, loss of less than 1.8 dB, and phase imbalance below 8°, over the complete ultrawideband bandwidth.
Optimization of renewable energy based hybrid energy system using evolutionary computational techniques
Adefarati, T.; Potgieter, S.; Sharma, G.; Bansal, Ramesh C.; Onaolapo, A.K.; Borisade, S.G.; Oloye, A.O. (Springer, 2025-02-11)
The sudden increase in global energy demandefor renewable energy resources. The global transition to renewable energy has emphasized the need for efficient, sustainable and cost-effective hybrid renewable energy system in the conventional power system. This study focuses on the optimization of HRES witeeh the aim objective of improving energy efficiency, sustainability and affordability. The proposed HRES which consists of standby diesel generator, wind turbines, battery storage system and photovoltaic system is designed to satisfy energy demands while reducing dependency on fossil fuels. The optimization of the power system is implemented with the eevolutionary computation techniques governed by particle swarm optimization and genetic algorithm in the MATLAB environment to coordinate the optimal power flow among several components of HRES. The techniques present in this research are based on the optimization of the total cost of the system and cost of energy of DG/PV/WT/BSS hybrid energy system. The hybridization of WT, PV and BSS in a single power system provides uninterrupted power supply to consumers at minimum CT of $11399 and $10906 as well as minimum COE of $0.1369/kWh and $0.1316/kWh by using GA and PSO. The findings show that the computational time to solve the problem by PSO is significantly less than that provided by GA. The optimal configuration has 72 PV panels (8.64 kW), 1 unit of WT (3 kW) and 76 battery systems (159.6 kWh) with computational time of 0.146831 s. The outcomes of the study demonstrate that HRES is a cost-effective solution to satisfy the power demand of the selected location and other regions based on similar meteorological data. The results obtained from the study align with Sustainable Development Goals by promoting clean energy access and fostering sustainable infrastructure development.
Stochastic energy management operation strategy for high penetrated grid connected solar with incorporation of battery storage system
Semwal, Gourav; Sharma, Sachin; Rawat, Tanuj; Sharma, Gulshan; Bansal, Ramesh C. (Springer, 2025-02-27)
The present distribution systems are heading towards smart distribution systems to attain large socio economic benefits. For achieving these benefits, the distribution system will include the practical aspect of the flexible modern technologies like renewable energy based distributed generators, demand response and battery energy storage system to manage load governance. There is source of uncertainty present in the non-dispatch able based distributed generations (DGs) that affects the operation schedule of distribution system. Therefore, this paper proposes a novel operation strategy for battery energy storage in coordination with uncertain large scale Photo-voltaic system based DG for distribution system. The optimal discharging and charging plans of battery energy storage for accommodation of uncertain photovoltaic are condition to the constraint like nodal power balance, feeder current limit and node voltage limit etc. Grey wolf optimization algorithm (GWO) is developed for analyzing the impact of multiple battery energy storage strategies, for controlling the demand deviation and node voltage of the distribution system. GWO algorithm is investigated on IEEE-33 bus radial distribution systems. The efficacy of the result shows the successful achieving the promised voltage profile and grid demand deviation profile.
MHealth technologies in voice disorders : a scoping review
Du Toit, Linette; Du Toit, Maria; Vermeulen, Rouxjeanne; Swanepoel, De Wet; Myburgh, Hermanus Carel; Patel, Rita; Van der Linde, Jeannie (Elsevier, 2026)
BACKGROUND : Technological advancements in healthcare offer the potential to improve patient outcomes, clinician productivity, and access to care. Evidence on their availability, clinical application, and long-term effectiveness in voice disorders remains unclear, highlighting the need for a comprehensive scoping review. AIM : To map and describe existing evidence on the use of mobile health (mHealth) technologies for the early detection, assessment, and treatment of voice disorders. METHODS : A scoping review was conducted according to the Joanna Briggs Institute (JBI) framework and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews Checklist (PRISMA-ScR) to ensure a comprehensive and systematic approach. RESULTS : Eighty-four studies were included, predominantly published between 2016 and 2025 and conducted in mostly high-income countries. Most focused on adult populations (67%) and the use of smartphones (51%) or telehealth platforms (19%). The mHealth solutions primarily targeted neurological (27%) and functional voice disorders (20%) and have demonstrated feasibility, accessibility, and potential for early detection, monitoring, and treatment. Most studies (31%) relied on acoustic assessments, while only 4% used gold-standard laryngeal imaging techniques, such as stroboscopy or endoscopy. CONCLUSION : mHealth technologies have the potential to enhance accessibility, equity, and cost-effectiveness in voice disorder care, particularly in underserved regions. Further research is needed to expand applications in early detection, diagnosis, and treatment, especially incorporating laryngeal imaging, as these solutions could potentially transform care into a preventative and globally sustainable model.
Towards a wearable skin tone responsive optical sensor
Ndiweni, Nomakhosi N.; Joubert, Trudi-Heleen (MDPI, 2025-09-22)
Melanin is one of the key light absorbers in skin and is responsible for the colour of the skin. This study evaluates the responsivity of different skin tones to white light within the visible spectral range of 300–700 nm on 12 participants. The results show that the peak amplitude of the reflected light signal decreased by 90% for darker skin tones, compared to 70% for lighter skin tones. There were also visible differences at the 460 nm and 570 nm wavelengths between the skin tones, suggesting that the standard one-glove-fits-all pulse oximeter might not be ideal.
Veterinary blood oxygen detection
Mpala, Kimberly; Joubert, Trudi-Heleen (MDPI, 2025-11-28)
A multimodal sensor was developed to record dissolved oxygen, L*a*b* colour, temperature, and pH. This work builds on an existing model that correlates blood oxygen saturation with L*a*b* colour values. An L*a*b* colour sensor was constructed from an RGB sensor and validated against a commercial colourimeter. Sensor performance was confirmed using reference colours. Dissolved oxygen was measured with a screen-printed electrode and an analogue-to-digital converter. The results highlight potential for future optical determination of oxygen saturation, combined with electrochemical measurement of oxygen partial pressure, and compensation for pH and temperature.
Paired emitter–detector diode array for colorimetric detection of water treatment chemicals
Olivier, Duane; Joubert, Trudi-Heleen (MDPI, 2025-09-13)
Optical spectroscopy is a versatile analytical technique with a diverse range of applications. Point-of-need systems are required to be affordable, miniaturized instruments that are easy to use. This paper proposes using an array of LEDs to create paired emitter detector diodes where commercial LEDs function as both a light source and detector. This system can measure the concentration of different chemicals via a set of discrete wavelengths. Calibration curves are presented for series of known concentrations of three water treatment chemicals using the K-matrix method. The spectral fingerprint identifies the chemical correctly with 99% accuracy using the Pearson correlation.
Use of machine learning to detect dangerous level of coal mine methane (CMM) concentrations during underground mining operations
Mooroogen, Rubeshen; Ayomoh, Michael Kweneojo (MDPI, 2025-11-07)
Underground coal mining is considered to be a highly dangerous activity and has been responsible for large amounts of accidents, causing the death of many mine workers. One of the factors responsible for the fatal aspect of underground coal mining is the presence and accumulation of toxic gases during underground mining operations. This paper focused its investigation specifically on coal mine methane (CMM), which is released as a result of the extraction of coal and the disturbance inflicted to surrounding rock formations during deep mining operations. Methane is considered a highly dangerous gas as it holds the capacity to cause explosions due to its highly inflammable nature. It can also displace oxygen, which eventually leads to asphyxiation. This research was based on the use of machine learning models to successfully predict dangerous concentrations of methane over the authorized threshold. Those predictions were made from a dataset containing information on the temperature, airflow, humidity, pressure and methane concentration in an underground coal mine. The temperature, airflow, humidity and pressure measurements were recorded by a series of sensors, namely anemometers and component sensors THP2/93. Three machine learning classification models were implemented and compared, with the objective to find the best model to predict and detect dangerous levels of coal mine methane. The models that were investigated included naïve Bayes, logistic regression and artificial neural networks (ANNs). This paper concludes with an engineering decision matrix that illustrates the precision of these models in predicting and detecting dangerous levels of methane concentrations in underground mines. Furthermore, recommendations for capacity improvement towards successfully predicting and detecting dangerous levels of coal mine methane from an artificial intelligence’s perspective are provided.
Electrode thickness optimization in color-selective inkjet-printed photosensitive organic field-effect transistors
Steger , Christoph; Tunc, Ali Veysel; Rainer, Christian; Karakaya, Ozan; Mager, Dario; Preciado, Luis Ruiz; Joubert, Trudi-Heleen; Lemmer, Uli; Hernandez-Sosa, Gerardo (MDPI, 2025-09-24)
This work introduces a general solution for printing wavelength-selective bulk-heterojunction photosensitive organic field effect transistors (PS-OFETs) by addressing electrode thickness variation and the feasibility of color selectivity in detecting incident light. The inkjet-printed silver electrode thickness was varied from 125 to 950 nm by multilayer printing. PIF, IDFBR, and ITIC-4F were chosen as the active semiconductor materials with complementary optical absorption. Results indicate that PS-OFETs exhibit the best functionality at an electrode thickness of approximately 325 nm and an active material combination with PIF:IDFBR (1:1). For the 540 nm wavelength, a responsivity of 55 mAW−1 was obtained. This is four-fold higher than the photoresponse obtained at 700 nm. .
Microfabrication of an e-QR code sensor display on a flexible substrate
Raju, Asha Elizabeth; Laue, Heinrich Edgar Arnold; Joubert, Trudi-Heleen (MDPI, 2025-09-19)
Electronic quick response (e-QR) codes provide access to real-time sensor data using smartphone readers and internet connectivity. Printed electronics and hybrid integration on flexible substrates is a promising solution for wide-scale and low-cost deployment of sensor systems. This paper presents a 21 × 21-pixel e-QR display implemented on black Kapton using hybrid additive and subtractive microfabrication techniques. The process flow for the double-sided circuit allows for layer alignment using multiple fiducial markers. The steps include inkjet printing of tracks on both sides of the substrate, laser-cut via holes, stencil-aided via filling, solder paste dispensing, and final integration of discrete surface-mount components by semi-automatic pick-and-place.
In-plane thermoelectric characterisation of PEDOT: PSS films with inkjet-printed test structures
Msomi, Promise N.; Joubert, Trudi-Heleen (MDPI, 2025-09-11)
A rapid screening method to identify suitable candidate inks for printed electronics applications is necessary. Herein, we investigate the in-plane thermoelectric properties of PEDOT:PSS for energy harvesting applications on human skin using silver nanoparticle inkjet-printed test structures. The in-plane electrical and thermal conductivity are measured. The Seebeck coefficient, ZT figure of merit, and power factor are consequently determined. PEDOT:PSS films resulted in low-efficiency thermoelectric properties at 293 K to 313 K and demonstrated a correlation between film thickness and in-plane thermoelectric properties. This study demonstrates that the test structures enable generalisable characterisation of thin-film inkjet-printable materials for thermoelectric purposes.
Data analysis and modelling of a sodium salt matrix with low-cost impedance spectroscopy
De Beer, Dirk Johannes; Joubert, Trudi-Heleen (MDPI, 2025-09-12)
This study investigates the use of impedance spectroscopy for fingerprinting aqueous salt solutions. By analysing solutions containing Sodium Nitrate (NaNO3), Sodium Sulphate (Na2SO4), and their mixtures, we explore how impedance data can potentially be used to distinguish between different salt compositions and concentrations. Our findings demonstrate the potential of this method for precise solution characterisation as well as highlighting the benefits of access to low-cost impedance analysers such as the one used for this investigation. More research and data is required to fully realise the potential of this approach.
Intelligent optimisation for onshore wind farm battery energy storage systems
Gwabavu, Mandisi; Bansal, Ramesh C.; Owolawi, P.A. (Elsevier, 2026-05)
Intelligent optimisation is essential for enhancing the performance and stability of onshore wind farms integrated with Battery Energy Storage Systems (BESSs), optimising turbine positioning, energy storage, and output to ensure efficient operation and sustained grid stability. Notwithstanding considerable advances in intelligent optimisation, there has been limited success in effectively integrating the diverse impacts of wind conditions and BESS state of health (SOH) while ensuring grid stability. This study investigates the application of intelligent optimisation for integrating BESS with onshore wind farms to augment energy storage capacity and ensure grid reliability. The proposed model incorporates a neural network (NN)-based inverse model, Bayesian optimisation, Gaussian Process Regression (GPR), and Reinforcement Particle Swarm Optimisation (RPSO) to enhance wind energy production while providing accurate estimates of system performance. Significant results from a case study of an onshore wind farm in South Africa and a 138 MW BESS demonstrate that the proposed intelligent optimisation model integrating NN-based inverse modelling, Bayesian-optimised GPR for SOH estimation, and RPSO enhance overall wind farm efficiency by 15-20% and reduce energy storage management errors by up to 10%. The proposed model is validated using NASA’s battery degradation datasets, achieving an RMSE of less than 1% in SOH estimation. These results highlight the potential of intelligent optimisation to enhance wind farms' performance, reliability, and stability integrated with battery energy storage systems, offering a scalable solution for renewable energy management. HIGHLIGHTS • Proposes a novel intelligent optimisation model integrating NN, GPR, Bayesian optimisation, and RPSO. • Enhances wind farm efficiency by 15–20% and reduces storage errors by up to 10%. • Achieves highly accurate battery SOH estimation with RMSE below 1%. • Improves grid stability and energy reliability in wind-BESS integration. • Validated using real South African wind farm data and NASA battery datasets.
Decoupled space vector modulation model and maximum modulation index analysis for nine-switch converters
Wang, Yi; Qin, Hao; Deng, Yan; Kumar, Abhishek; Wu, Jiande; Bansal, Ramesh C.; Naidoo, Raj M. (Institute of Electrical and Electronics Engineers, 2026-04)
Nine-switch converter (NSC) has been developed as a compact solution for dual-output applications, but its switch-sharing architecture introduces inherent coupling problems between output ports that poses modulation difficulties. This article establishes a decoupled space vector modulation (SVM) model for NSC to address this fundamental challenge. Through analytical reconstruction of the conventional SVM model and decomposition of reference vectors, the proposed modeling method achieves decoupled modulation and independent port control while reducing the complexity of SVM algorithm. Based on the proposed decoupled model, a comprehensive mathematical analysis of the maximum modulation index is derived and the analytical expression formula is obtained, explicitly revealing the phase-dependent linear modulation range of NSC. The performance of the proposed SVM method is validated through simulation and a 400 V, 3 kW experimental prototype and the experiment results exhibit strong agreement with the theoretical analysis of maximum modulation indices.
Constant voltage-induced intermittent discharge behavior of metal contaminant degradation on the surface of GIS insulators
Zhang, Guozhi; Chen, Xu; Ye, Xianming; Zhang, Xiaoxing (Institute of Electrical and Electronics Engineers, 2026)
In recent years, there have been many breakdown discharge accidents caused by intermittent discharge in GIS, and there is no effective method to obtain the change characteristics of intermittent discharge defects under constant voltage to breakdown discharge. Based on this, this paper further studies the deterioration and discharge characteristics of intermittent discharge defects in surface metal pollution of GIS insulators under constant voltage, establish a GIS insulator surface metal contamination intermittent discharge degradation test platform. The discharge quantity and UHF signal changes of intermittent discharge defects caused by metal contamination on the surface of insulators under constant voltage were investigated separately under different temperatures and ultra-long discharge times, and realizes the rapid evolution of metal pollution intermittent discharge into breakdown discharge under constant voltage for the first time. The increase in temperature will cause the defect to evolve from an intermittent discharge state to a stable discharge state; With the increase of time, the discharge amount and UHF signal amplitude increase significantly with the development of pollution deterioration, from intermittent discharge to stable discharge, the discharge intensity suddenly surges at about 200min, and finally rapidly develops into breakdown discharge, which is consistent with the breakdown phenomenon observed in the GIS equipment of Wuhu station.
In silico prediction method for plant nucleotide-binding leucine-rich repeat- and pathogen effector interactions
Fick, Alicia; Fick, Jacobus Lukas Marthinus; Swart, Velushka; Van den Berg, Noelani (Wiley, 2025-04)
Plant Nucleotide-binding leucine-rich repeat (NLR) proteins play a crucial role in effector recognition and activation of Effector triggered immunity following pathogen infection. Genome sequencing advancements have led to the identification of a myriad of NLRs in numerous agriculturally important plant species. However, deciphering which NLRs recognize specific pathogen effectors remains challenging. Predicting NLR-effector interactions in silico will provide a more targeted approach for experimental validation, critical for elucidating function, and advancing our understanding of NLR-triggered immunity. In this study, NLR-effector protein complex structures were predicted using AlphaFold2-Multimer for all experimentally validated NLR-effector interactions reported in literature. Binding affinities- and energies were predicted using 97 machine learning models from Area-Affinity. We show that AlphaFold2-Multimer predicted structures have acceptable accuracy and can be used to investigate NLR-effector interactions in silico. Binding affinities for 58 NLR-effector complexes ranged between -8.5 and -10.6 log(K), and binding energies between -11.8 and -14.4 kcal/mol-1, depending on the Area-Affinity model used. For 2427 "forced" NLR-effector complexes, these estimates showed larger variability, enabling identification of novel NLR-effector interactions with 99% accuracy using an Ensemble machine learning model. The narrow range of binding energies- and affinities for "true" interactions suggest a specific change in Gibbs free energy, and thus conformational change, is required for NLR activation. This is the first study to provide a method for predicting NLR-effector interactions, applicable to all pathosystems. Finally, the NLR-Effector Interaction Classification (NEIC) resource can streamline research efforts by identifying NLRs important for plant-pathogen resistance, advancing our understanding of plant immunity.
Intelligent optimisation for sustainable development of onshore wind farm battery energy storage systems : a systematic review
Gwabavu, Mandisi; Bansal, Ramesh C.; Shakantu, Winston; Nyandeni, Lethaba; Owolawi, Pius Adewale.; Madwe, Mziwendoda C. (Elsevier, 2026-03)
The global shift toward renewable energy is driven by the urgent need to combat climate change and achieve net-zero emissions, with wind power emerging as a key pillar due to its scalability and cost-effectiveness amid rising energy demands and fossil fuel volatility. This systematic review investigates the role of intelligent optimisation techniques in facilitating the sustainable development of onshore wind farms integrated with Battery Energy Storage Systems (BESS). It synthesises advanced methods, including machine learning, evolutionary algorithms, and hybrid meta-heuristics, to enhance reliability, efficiency, and lifecycle performance of wind-BESS systems, addressing wind intermittency challenges in low-carbon transitions. Employing a systems-based approach and the PRISMA methodology, the review analyses 1295 studies, of which 152 meet eligibility criteria, encompassing technological innovation, facilities management (FM), and social, economic, and environmental factors. The primary contribution is the integration of FM perspectives into intelligent optimisation, proposing a novel framework that links predictive analytics, asset management, and sustainability metrics. Results highlight the efficacy of intelligent optimisation in improving grid stability, reducing costs, and enhancing resilience. The review proposes a conceptual framework that aligns optimisation with global sustainability goals and calls for empirical validation to guide policy, investment, and practice, particularly in developing economies. HIGHLIGHTS • This review's bibliometric analysis evaluated 1295 studies, with 152 meeting strict eligibility criteria. • Life cycle stages of intelligent optimisation for sustainable development of an onshore wind farm BESS. • Identifying critical factors for the sustainable development of an onshore wind farm BESS. • The integration of FM perspectives into intelligent optimisation of wind farm BESS. • A theoretical and conceptual framework guiding intelligent optimisation in sustainable development of onshore wind farm BESS.
Simultaneous sizing of a photovoltaic system and compressed air energy storage in a microgrid
Kalala, Tshilumba; Mbukani, Mwana Wa Kalaga (Elsevier, 2026-05)
The integration of Compressed Air Energy Storage (CAES) with photovoltaic (PV) systems, complemented by grid interconnection capabilities and diesel generator backup, represents an advanced approach to sustainable microgrid design for future energy systems. In this study, a multi-objective optimization model for sizing PV-CAES systems is formulated as a Mixed-Integer Nonlinear Programming (MINLP) problem with two primary objective functions: (1) minimization of total system investment costs (CAPEX) and operational costs (OPEX), and (2) enhancement of system reliability and maximization of RE penetration. The Augmented -constraint method is applied to solve this multi-objective optimization problem by incorporating the reliability and RE penetration objectives as inequality constraints, while maintaining cost minimization as the overall optimization goal. In application to a case study of a South African commercial building, the optimized design saves annual operational costs by 35.2% and achieves 41.5% penetration of RE and 2.4% increase in reliability compared with conventional designs. The results demonstrate the success of the framework in providing economically viable PV-CAES configurations that simultaneously enhance sustainability and system reliability via comprehensive mathematical optimization. HIGHLIGHTS • An optimization model is proposed for PV and CAES sizing in a microgrid. • Previous probabilistic methods are improved with mathematical optimization. • The model includes energy balance and storage dynamics constraints. • The approach reduces system cost and improves energy reliability. • Results support better microgrid planning with high renewable penetration.
An emulated dynamic framework for evaluating metaheuristic-based load balancing techniques in edge computing networks
Molokomme, Daisy Nkele; Onumanyi, Adeiza James; Abu-Mahfouz, Adnan Mohammed (MDPI, 2026-03)
Edge computing (EC) has emerged as a paradigm to support computation-intensive Internet of Things (IoT) applications by enabling task offloading to nearby servers. Despite its potential, the inherent heterogeneity of edge resources and the dynamic, unpredictable nature of task arrivals present significant challenges for designing and evaluating effective load balancing strategies. Traditional evaluation methods are limited as follows: physical testbeds lack scalability and flexibility, while abstract simulators often oversimplify network behavior, failing to capture realistic system dynamics. To address these limitations, we present an emulated dynamic edge computing framework (EDECF) designed for evaluating load balancing schemes in EC networks. First, we developed dedicated service models for each EC node within the EDECF and implemented them using the common open research emulator (CORE) platform, thereby providing a scalable, flexible, and realistic environment for testing optimization strategies. Second, we introduced a robust fitness function that explicitly models latency, queue stability, and fairness for metaheuristic-based load balancing under dynamic edge conditions. To assess its effectiveness, this function was incorporated and tested using the following methods: the particle swarm optimization, genetic algorithm, differential evolution and simulated annealing-based load balancing algorithms. In addition, baseline methods such as the round robin and shortest queue techniques were also deployed to demonstrate the framework’s capacity to facilitate rigorous analysis in heterogeneous and time-varying scenarios. Overall, results are presented to demonstrate EDECF’s capability to emulate realistic workloads, capture resource variability at the edge, and support comprehensive evaluation of algorithmic performance across diverse network settings. Thus, this work aims to establish a practical and extensible foundation for researchers and practitioners to design, test, and optimize load balancing strategies in EC environments.
Artificial intelligence based learning methods for the automatic tuning of fixed-parameter MIMO PID controllers for industrial applications : a review and comparison
Van Niekerk, Jonathan Anson; Le Roux, Johan Derik; Craig, Ian Keith (Elsevier, 2026-05)
This paper reviews and compares artificial intelligence (AI) methods for the automatic tuning of multi-input-multi-output (MIMO) proportional-integral-derivative (PID) controllers in industrial process applications. The study focuses on fixed-parameter PID tuning and introduces a generalised procedure that unifies diverse AI methods within a single autotuning framework. A Pareto-front-based weighting strategy is proposed to balance performance and actuator usage, enabling fair comparison of tuning outcomes across different algorithms. Within this framework, three representative approaches, particle swarm optimisation (PSO), proximal policy optimisation (PPO), and Bayesian optimisation (BO), are implemented and evaluated against the defining criteria of an ideal autotuner: versatility, global optimality, data efficiency, and safety. The analysis bridges computational intelligence and machine learning perspectives, providing a structured benchmark for assessing AI-based tuning performance. Results show that all AI-based tuners successfully identify high-performing controller parameters for multivariable nonlinear systems, confirming their applicability to industrial processes. Among them, BO achieves the best overall performance, offering superior convergence speed and data efficiency through surrogate-driven optimisation. By maximising information gained from each plant trial, BO provides a safe, robust, and computationally efficient tuning method ideally suited to practical industrial deployment.

Browse

Recent Submissions