Zenghui Wang

• PhD
• Professor (Full) at University of South Africa

This research presents a novel approach to multilayer perceptron's (MLP) hyper-parameter optimization in solving learner migration problems in Limpopo, South Africa. While acknowledging the presence of various hyper-parameter optimization techniques, their applicability, strengths, and limitations differ. Our approach utilizes meta-heuristics, offe...

Extension of the feeders and high power demand have resulted in more substantial power dissipation and reduced voltage decline. Therefore, minimizing power loss and enhancing the voltage profile are imperative for efficient energy distribution. This research introduces a novel approach called the Hybrid Archimedes Optimization Algorithm (HAOA), whi...

Since the effusion of Industry 4.0 (I40) and smart manufacturing, predictive maintenance (PdM) has become critical to prevent severe system breakdowns and costly production downtime in various industries. Several state-of-the-art Artificial Intelligence (AI) approaches, such as machine learning models (ML), empower the PdM design concept to produce...

The South African Education Management Information Systems (EMIS) hosts longitudinal data on school inventory, learners, and educators. One of the most prevailing and yet ignored phases in machine learning is Feature Selection (FS). Neglecting this phase can adversely impact the outcome of the machine-learning exercise. This study seeks to explore...

The South African Education Management Information Systems (EMIS) hosts longitudinal data on school inventory, learners, and educators. One of the most prevailing and yet ignored phases in machine learning is Feature Selection (FS). Neglecting this phase can adversely impact the outcome of the machine-learning exercise. This study seeks to explore...

The detection of human emotions from speech signals remains a challenging frontier in audio processing and human-computer interaction domains. This study introduces a novel approach to Speech Emotion Recognition (SER) using a Dendritic Layer combined with a Capsule Network (DendCaps). A Convolutional Neural Network (NN) and a Long Short-Time Neural...

The effectiveness of deep learning networks in detecting small objects is limited, thereby posing challenges in addressing practical object detection tasks. In this research, we propose a small object detection model that operates at multiple scales. The model incorporates a multi-level bidirectional pyramid structure, which integrates deep and sha...

Credit risk prediction is a crucial task for financial institutions. The technological advancements in machine learning, coupled with the availability of data and computing power, has given rise to more credit risk prediction models in financial institutions. In this paper, we propose a stacked classifier approach coupled with a filter-based featur...

Intrusion detection systems play a critical role in the mitigation of cyber-attacks on the Internet of Things (IoT) environment. Due to the integration of many devices within the IoT environment, a huge amount of data is generated. The generated data sets in most cases consist of irrelevant and redundant features that affect the performance of the...

The capabilities of artificial intelligence (AI) and deep Learning are increasing rapidly with the increasing computing power and specialized microprocessors. A very interesting architecture, generative adversarial networks (GANs), is at the forefront of innovation. Some examples of what GAN networks are used for are text-to-image translation, imag...

The electrical power system (EPS) has been heavily stressed due to high load demand. It operates close to the total capacity limits, resulting in voltage instability that can lead to voltage collapse. In this regard, incorporating flexible alternating current transmission system (FACTS) devices and renewable energy sources (RESs) to obtain the opti...

This study compares the performance of stream clustering algorithms (DenStream, CluStream, ClusTree) on Massive Online Analysis (MOA) using synthetic and real-world datasets. The algorithms are compared in the presence on noise level [0%, 10%, 30%] on the synthetic data. DenStream epsilon parameter was tune to 0.01 and 0.03 to improve its performan...

Parameter specification remains a difficult task in data stream clustering as density-based algorithms hyperparameters tuning to their optimal values are often difficult to determine. This paper investigates the sensitivity of parameter tuning on DenStream, a data stream clustering algorithm. The effects on different noise levels are evaluated for...

System identification is a very critical process in a control system aiming to determine the system dynamics of practical processes before devising an adequate control approach. Several metaheuristic approaches have been used for the parametric estimation of nonlinear systems, ranging from classical to new-generation metaheuristics. Nevertheless, t...

The bat algorithm (BA) is a population-based optimization that mimics the echolocation of microbats when looking for prey and avoiding obstacles. BA is a unique algorithm that gives fast convergence and optimum solutions to a problem. Reactive power dispatch (RPD) plays a vital role in the control and operation of the power system and is part of th...

Weak bus identification in power system networks is crucial for planning and operation since most generators operate close to their operating limits, resulting in generator failures. This work aims to identify the critical/weak node and reduce the system’s actual power losses. The line stability index (\({L}_{mn}\)) and fast voltage stability index...

Identifying the weak buses in power system networks is crucial for planning and operation since most generators operate close to their operating limits, resulting in generator failures. This work aims to identify the critical/weak node and reduce the system’s power loss. The line stability index () and fast voltage stability index (FVSI) were used...

The distillation process plays an essential role in the petrochemical industry. However, the high-purity distillation column has complicated dynamic characteristics such as strong coupling and large time delay. To control the distillation column accurately, we proposed an extended generalized predictive control (EGPC) method inspired by the princip...

The voltage control problem due to bidirectional power flows is more apparent when heterogeneous distributed generation systems (DGS) are integrated into the grid. In this paper, a novel method of voltage control in distributed generation systems based on a reinforcement learning technique is proposed. DGS incorporating renewable energy resources a...

Marine sensors are highly vulnerable to illegal access network attacks. Moreover, the nation’s meteorological and hydrological information is at ever-increasing risk, which calls for a prompt and in depth analysis of the network behavior and traffic to detect network attacks. Network attacks are becoming more diverse, with a large number of rare an...

Fractional order PID tuning has surfaced as a substitute controller for PIDs offering better regulatory system operation with an additional increase in estimation parameters. Over the years, the three main representative metaheuristics global optimisation techniques (Genetic algorithm, differential evolution and particle swarm optimisation) have be...

Electricity has become one of the most essential components of establishing a quality standard of living in any country. Consequently, considerable work has been focused on designing a sophisticated load frequency control (LFC) system. However, in light of limited resources and real-world challenges, computationally based control algorithms that ar...

Wind power is the most promising renewable energy for its rich resources, low cost, and cleanliness. However, the intermittency of wind power would put the safety of the power system at risk. An effective method to solve this problem is making the power generation scheduling through wind power forecast. Due to the volatility and complex temporal de...

This research presents a new key information extraction algorithm from shopping receipts. Specifically, we train semantic, visual and structural features through three deep learning methods, respectively, and formulate rule features according to the characteristics of shopping receipts. Then we propose a multi-class text classification algorithm ba...

In this article, we present the recognition of nonintrusive disaggregated appliance signals through a reduced dataset computer vision deep learning approach. Deep learning data requirements are costly in terms of acquisition time, storage memory requirements, computation time, and dynamic memory usage. We develop our recognition strategy on Siamese...

In the real applications, we found that it is difficult to achieve good control performance through manually tuning proportional–integral (PI) parameters of phase locked loop (PLL) and speed-loop of Luenberger observer (LO) for the PMSM sensorless control system. Therefore, this paper is to use the particle swarm optimization (PSO) algorithm to opt...

This paper reports a method to derive a novel class of 3D generalized thermostatted oscillators from a simple damped harmonic oscillator. Its detailed procedure is obtained through a mathematical derivation. Then, we propose an example system to show the effectiveness of the method. Furthermore, the numerical analysis is performed to investigate it...

There is an ever-increasing risk of illegal access-induced Network Intrusion (NI), which calls for prompt detection of illegal network behavior through profound Network Traffic (NT) analyses. However, current intrusion detection methods are limited in accuracy due to insufficient data standardization. This paper puts forward a deoxyribonucleic acid...

Memristors are usually introduced into neuron models as neural synapses to investigate firing activities. In this paper, a novel generic memristor with smooth cosine memductance is proposed, and its dynamic characteristic concerning multistability, which is completely different from any known memristors, is analyzed and validated by numerical and P...

Face recognition is the problem of identifying or recognizing individuals in an image. This paper investigates a possible method to bring a solution to this problem. The method proposes an amalgamation of Principal Component Analysis (PCA), K-Means clustering, and Convolutional Neural Network (CNN) for a face recognition system. It is trained and e...

Loss of selection pressure in the presence of many objectives is one of the pertinent problems in evolutionary optimization. Therefore, it is difficult for evolutionary algorithms to find the best-fitting candidate solutions for the final Pareto optimal front representing a multi-objective optimization problem, particularly when the solution space...

As computer networks keep growing at a high rate, achieving confidentiality, integrity, and availability of the information system is essential. Intrusion detection systems (IDSs) have been widely used to monitor and secure networks. The two major limitations facing existing intrusion detection systems are high rates of false-positive alerts and lo...

The chaotic system with complicated dynamical behaviors has high potential in practical applications. This paper reports a simple smooth 3D dynamical system that is derived from the Nosé-Hoover oscillator and has rich dynamical behaviors, such as fold-Hopf bifurcation, saddle-node bifurcation, transient chaos, and conservative chaos (hidden attract...

The prediction of host human miRNA binding to the SARS-COV-2-CoV-2 RNA sequence is of particular interest. This biological process could lead to virus repression, serve as biomarkers for diagnosis, or as potential treatments for this disease. One source of concern is attempting to uncover the viral regions in which this binding could occur, as well...

The object detector based on deep learning has received extensive attention, but the high computational cost has become an obstacle to its large-scale application. It is a great challenge for object detection to further reduce the hardware requirements on the premise of ensuring high detection accuracy. We propose a one-stage lightweight object det...

Multiobjective evolutionary algorithm based on decomposition (MOEA/D) has attracted a lot of attention since it can handle multiobjective problems (MOP) with a complicated Pareto front. The procedure involves decomposing a MOP into single subproblems, which are eventually optimized simultaneously based on the MOP neighborhood information. However,...

Distributed generation (DG) plays a vital role in electrical power networks. However, power loss reduction, voltage profile improvement, friendly environment, and reliability are all benefits of DG units. In this research work, a worthwhile methodology is recommended for optimal allocation of traditional (gas turbine) and renewable energy sources t...

Abstract This paper proposes a new hybridised approach comprising the flower pollination algorithm and pathfinder algorithm (FPAPFA), in order to address optimisation problems and for load frequency control system. Although the FPA is a popular algorithm that has been widely used in diverse applications, its implementation is met with the tendency...

In this paper, the suitability of audio features for application in speech-music discrimination was evaluated to select a feature set that produces high mean accuracy in the classification algorithm, while also reducing the total feature space. The first four standardized moments of twelve audio features were evaluated namely the mean, variance, sk...

Wireless Sensor Networks (WSNs) create various security threats such as application variance in different sectors along with the model of cryptographic primitive and necessity. Although modernistic evolution, the skillful utilization of Elliptic Curve Cryptography (ECC) for WSNs is a very progressive investigation topic and approaches to reduce the...

It is important to control the ship course in complicated ocean environment. In this paper, a Grey Wolf Optimization (GWO) based Active Disturbance Rejection Control (ADRC) tuning method is proposed in the application of the ship course. Here, GWO is used to tune the parameters of ADRC. To validate the performance of the proposed method, some simul...

The recent advances of e-commerce and e-payment systems have sparked an increase in financial fraud cases such as credit card fraud. It is therefore crucial to implement mechanisms that can detect the credit card fraud. Features of credit card frauds play important role when machine learning is used for credit card fraud detection, and they must be...

The industrial manufacturing sector is undergoing a tremendous revolution moving from traditional production processes to intelligent techniques. Under this revolution, known as Industry 4.0 (I40), a robot is no longer static equipment but an active workforce to the factory production alongside human operators. Safety becomes crucial for humans and...

The high-purity distillation column system is strongly nonlinear and coupled, which makes it difficult to control. Active Disturbance Rejection Control (ADRC) has been widely used in distillation systems, but it has limitations in controlling distillation systems with large time delays since ADRC employs ESO and feedback control law to estimate the...

The thermostatted system is a conservative system different from Hamiltonian systems, and has attracted much attention because of its rich and different nonlinear dynamics. We report and analyze the multiple equilibria and curve axes of the cluster-shaped conservative flows generated from a generalized thermostatted system. It is found that the clu...

Inspired by the structure of the single-clew-shaped conservative chaotic flows generated from the Nosé–Hoover oscillator, the possible local dynamic behaviors of the thermostatted oscillator are discovered near the axis of the clew-shaped chaotic flows. Based on the formalism of the port-controlled Hamiltonian system, we propose a variant of the No...

αBB is an elegant deterministic branch and bound global optimisation that guarantees global optimum convergence with minimal parameter tuning. However, the method suffers from a slow convergence speed calling for computational improvements in several areas. The current paper proposes hybridising the branch and bound process with particle swarm opti...

The advance in technologies such as e-commerce and financial technology (FinTech) applications have sparked an increase in the number of online card transactions that occur on a daily basis. As a result, there has been a spike in credit card fraud that affects card issuing companies, merchants, and banks. It is therefore essential to develop mechan...

The Industrial Internet of things (IIoT), the implementation of IoT in the industrial sector, requires a deterministic, real-time, and low-latency communication response for its time-critical applications. A delayed response in such applications could be life-threatening or result in significant losses for manufacturing plants. Although several mea...

Compared with dissipative chaos, conservative chaos is more suitable for information encryption since it does not have attractors, therefore, it is important to construct different complicated conservative chaotic systems. This paper reports an effective approach to construct a class of 4D conservative systems based on the formalism of a generalize...

The high-speed anti-ship missile might encounter a challenge in the terminal phase if there are mixed targets, including false and true ones. There is only a quite short time to take action after the onboard radar can distinguish which targets are true or false. In this study, a tunable acceleration feedback gain is used to realize a unique integra...

The industrial manufacturing sector is currently undergoing a tremendous revolution moving from traditional production processes to intelligent techniques. Under this revolution, known as Industry 4.0 (I40), a robot is no longer a static equipment but an active workforce to the factory production alongside human operators. Safety becomes crucial fo...

As the world’s population grows and energy demand increases, it is necessary to increase the scale of the electrical system, which is more complicated. Consequently, adopting automatic generation control (AGC) scheme to meet the demand becomes inevitable. In this article, the fusion of flower pollinated algorithm (FPA) and pathfinder algorithm (PFA...

In this paper, a modified Sprott-C chaotic system is proposed based on Kolmogrov model, which shows rich dynamic behaviors, especially, the system divergence is related to the variables. To quantitatively evaluate the influence of variable divergence on phase space volume, the ultimate bound and equilibrium point of the system are analyzed and two...

The influence of non-coding RNAs, such as lncRNAs (long non-coding RNAs) and miRNAs (microRNAs), is undeniable in several diseases, for example, in the formation of neoplasms and cancer scenarios. However, there are challenges due to the scarcity of validated datasets and the imbalance in the data. We found that the research of associations between...

In this paper, a new type of non-volatile locally active memristor with bi-stability is proposed by injecting appropriate voltage pulses to realize a switching mechanism between two stable states. It is found that the memristive parameters of the new memristor can affect the local activity, which has been rarely reported, and this phenomenon is exp...

The relationship between smart devices and human beings is one of the research hotspots of the Fourth Industrial Revolution (4IR). In this regard, we explored the practical relationship between the 3D electric field components measured by the smart 3D atmospheric electric field apparatus (AEFA) and the thunderstorm activity from the perspective of...

Chaotic systems have high potential for engineering applications due to their extremely complex dynamics. In the paper, a five-dimensional (5D) Kolmogorov-like hyperchaotic system is proposed. First, the hyperchaotic property is uncovered, and numerical analysis shows that the system displays the coexistence of different kinds of attractors. This s...

The discriminative correlation filter-based tracking algorithms cannot correctly track the target if the target is occluded or out of view and reappears in the field of vision, and they cannot ensure the tracking model is updated correctly if the tracking information is not correct. In this paper, a robust correlation tracking algorithm is proposed...

One challenge in miRNA-genes-diseases interaction studies is that it is challenging to find labeled data that indicate a positive or negative relationship between miRNA and genes. The use of one-class classification methods shows a promising path for validating them. We have applied two one-class classification methods, Isolation Forest and One-cla...

Pseudo-random number generator (PRNG) has been widely used in digital image encryption and secure communication. This paper reports a novel PRNG based on a generalized Sprott-A system that is conservative. To validate whether the system can produce high quality chaotic signals, we numerically investigate its conservative chaotic dynamics and the co...

Medical data generated from hospitals are an increasing source of information for automatic medical diagnosis. These data contain latent patterns and correlations that can result in better diagnosis when appropriately processed. Most applications of machine learning (ML) to these patient records have focused on utilizing the ML algorithms directly,...

In this paper, a new type of non-volatile locally active memristor with bi-stability is proposed by injecting appropriate voltage pulses to realize a switching mechanism between two stable states. It is found that the memristive parameters of the new memristor can affect the local activity, which has been rarely reported, and this phenomenon is exp...

The prevailing competitive manufacturing industry calls for continuous customer satisfaction for business sustainability. With the emergence of the Industry 4.0 paradigm, product customization, which gives customers the means to personalized products to meet their needs, has become a strategy to increase companies’ value. High-tech manufacturing fi...

The mains signal is a complex fusion of various electrical equipment load signals in a building. In the non-intrusive load monitoring recognition, our main aim is to be able to extract as much load features as possible from the complex aggregate mains signal in a simpler way through a computer vision-based approach as opposed to the powers series s...

In recent years, Convolutional Neural Network (CNN) has been widely applied in speech/image/video recognition and classification. Although the results achieved are so impressive, CNN architecture is becoming more and more complex since CNN includes more layers to achieve better performance. In this paper, we developed a new CNN structure with sever...

Lung cancer is the second most common form of cancer in both men and women. It is responsible for at least 25% of all cancer-related deaths in the United States alone. Accurate and early diagnosis of this form of cancer can increase the rate of survival. Computed tomography (CT) imaging is one of the most accurate techniques for diagnosing the dise...

The near-space hypersonic aerodynamic glider has strong maneuverability in wide flight envelope. The glide is generally achieved in a smooth manner with no or weak altitude oscillations in the altitude. The maximum lift-to-drag ratio glide is a typical trajectory that can approximate the maximum glide range, which is a crucial indicator for a glide...

Automatic Voltage Regulator (AVR) system is one of the major devices broadly used in many industrial applications for regulating the voltage of the synchronous generator within its nominal values. Consequently, providing a suitable controller for the AVR system becomes a necessity to prevent instability and error in the system’s output response. St...

This paper presents the optimization of the fractional order proportional-integral-derivative (FOPID) controller with the hybrid particle swarm and grey wolf optimization (HPSGWO) algorithm in order to control the Automatic Voltage Regulator (AVR) system. To overcome the problem of premature convergence and enhance the efficiency of the proposed HP...

Algorithms are used to optimize both single and multi-objective system limits. This research aimed to detect the optimal location and size of the DGs, which can significantly minimize power loss and improve the stability of the voltage. The research uses binary particle swarm optimization and shuffled frog leap (BPSO-SLFA) algorithms for simulation...

This paper introduces novel concepts for accelerating learning in an off-policy reinforcement learning algorithm for Partially Observable Markov Decision Processes (POMDP) by leveraging multiple agents frame work. Reinforcement learning (RL) algorithm is considerably a slow but elegant approach to learning in an unknown environment. Although the ac...

In a smart home, the nonintrusive load monitoring recognition scheme normally achieves high appliance recognition performance in the case where the appliance signals have widely varying power levels and signature characteristics. However, it becomes more difficult to recognize appliances with equal or very close power specifications, often with alm...

Non-intrusive-load-monitoring (NILM) is normally based on power series analysis. In the load classification stage we use an image based deep convolutional neural network (DCNN) which is modelled on the biological visual cortex thereby achieving extremely high levels of object recognition and classification. However, the downsize to the DCNN is the...