Technologies

25 pages, 6413 KiB

Open AccessArticle

Power Tracking and Performance Analysis of Hybrid Perturb–Observe, Particle Swarm Optimization, and Fuzzy Logic-Based Improved MPPT Control for Standalone PV System

by Ali Abbas, Muhammad Farhan, Muhammad Shahzad, Rehan Liaqat and Umer Ijaz

Technologies 2025, 13(3), 112; https://doi.org/10.3390/technologies13030112 (registering DOI) - 8 Mar 2025

Viewed by 229

Abstract

The increasing energy demand and initiatives to lower carbon emissions have elevated the significance of renewable energy sources. Photovoltaic (PV) systems are pivotal in converting solar energy into electricity and have a significant role in sustainable energy production. Therefore, it is critical to [...] Read more.

The increasing energy demand and initiatives to lower carbon emissions have elevated the significance of renewable energy sources. Photovoltaic (PV) systems are pivotal in converting solar energy into electricity and have a significant role in sustainable energy production. Therefore, it is critical to implement maximum power point tracking (MPPT) controllers to optimize the efficiency of PV systems by extracting accessible maximum power. This research investigates the performance and comparison of various MPPT control algorithms for a standalone PV system. Several cases involving individual MPPT controllers, as well as hybrid combinations using two and three controllers, have been simulated in MATLAB/SIMULINK. The sensed parameters, i.e., output power, voltage, and current, specify that though individual controllers effectively track the maximum power point, hybrid controllers achieve superior performance by utilizing the combined strengths of each algorithm. The results indicate that individual MPPT controllers, such as perturb and observe (P&O), particle swarm optimization (PSO), and fuzzy logic (FL), achieved tracking efficiencies of 97.6%, 90.3%, and 90.1%, respectively. In contrast, hybrid dual controllers such as P&O-PSO, PSO-FL, and P&O-FL demonstrated improved performance, with tracking efficiencies of 96.8%, 96.4%, and 96.5%, respectively. This research also proposes a new hybrid triple-MPPT controller combining P&O-PSO-FL, which surpassed both individual and dual-hybrid controllers, achieving an impressive efficiency of 99.5%. Finally, a comparison of all seven cases of MPPT control algorithms is presented, highlighting the advantages and disadvantages of individual as well as hybrid approaches. Full article

19 pages, 13798 KiB

Open AccessArticle

RANFIS-Based Sensor System with Low-Cost Multi-Sensors for Reliable Measurement of VOCs

by Keunyoung Kim and Woosung Yang

Technologies 2025, 13(3), 111; https://doi.org/10.3390/technologies13030111 - 7 Mar 2025

Viewed by 297

Abstract

This study describes a sensor system for continuous monitoring of volatile organic compounds (VOCs) emitted from small industrial facilities in urban centers, such as automobile paint facilities and printing facilities. Previously, intermittent measurements were made using expensive flame ionization detector (FID)-type instruments that [...] Read more.

This study describes a sensor system for continuous monitoring of volatile organic compounds (VOCs) emitted from small industrial facilities in urban centers, such as automobile paint facilities and printing facilities. Previously, intermittent measurements were made using expensive flame ionization detector (FID)-type instruments that were impossible to install, resulting in a lack of continuous management. This paper develops a low-cost sensor system for full-time management and consists of multi-sensor systems to increase the spatial resolution in the pipe. To improve the accuracy and reliability of this system, a new reinforced adaptive neuro fuzzy inference system (RANFIS) model with enhanced preprocessing based on the adaptive neuro fuzzy inference system (ANFIS) model is proposed. For this purpose, a smart sensor module consisting of low-cost metal oxide semiconductors (MOSs) and photo-ionization detectors (PIDs) is fabricated, and an operating controller is configured for real-time data acquisition, analysis, and evaluation. In the front part of the RANFIS, interquartile range (IQR) is used to remove outliers, and gradient analysis is used to detect and correct data with abnormal change rates to solve nonlinearities and outliers in sensor data. In the latter stage, the complex nonlinear relationship of the data was modeled using the ANFIS to reliably handle data uncertainty and noise. For practical verification, a toluene evaporation chamber with a sensor system for monitoring was built, and the results of real-time data sensing after training based on real data were compared and evaluated. As a result of applying the RANFIS model, the RMSE of the MQ135, MQ138, and PID-A15 sensors were 3.578, 11.594, and 4.837, respectively, which improved the performance by 87.1%, 25.9%, and 35.8% compared to the existing ANFIS. Therefore, the precision within 5% of the measurement results of the two experimentally verified sensors shows that the proposed RANFIS-based sensor system can be sufficiently applied in the field. Full article

(This article belongs to the Topic Internet of Things Architectures, Applications, and Strategies: Emerging Paradigms, Technologies, and Advancing AI Integration)

► Show Figures

Figure 1

19 pages, 30651 KiB

Open AccessArticle

Comparative Evaluation of Commercial, Freely Available, and Open-Source Tools for Single-Cell Analysis Within Freehand-Defined Histological Brightfield Image Regions of Interest

by Filippo Piccinini, Marcella Tazzari, Maria Maddalena Tumedei, Nicola Normanno, Gastone Castellani and Antonella Carbonaro

Technologies 2025, 13(3), 110; https://doi.org/10.3390/technologies13030110 - 7 Mar 2025

Viewed by 132

Abstract

In the field of histological analysis, one of the typical issues is the analysis of single cells contained in regions of interest (i.e., ROIs). Today, several commercial, freely available, and open-source software options are accessible for this task. However, the literature lacks recent [...] Read more.

In the field of histological analysis, one of the typical issues is the analysis of single cells contained in regions of interest (i.e., ROIs). Today, several commercial, freely available, and open-source software options are accessible for this task. However, the literature lacks recent extensive reviews that summarise the functionalities of the opportunities currently available and provide guidance on selecting the most suitable option for analysing specific cases, for instance, irregular freehand-defined ROIs on brightfield images. In this work, we reviewed and compared 14 software tools tailored for single-cell analysis within a 2D histological freehand-defined image ROI. Precisely, six open-source tools (i.e., CellProfiler, Cytomine, Digital Slide Archive, Icy, ImageJ/Fiji, QuPath), four freely available tools (i.e., Aperio ImageScope, NIS Elements Viewer, Sedeen, SlideViewer), and four commercial tools (i.e., Amira, Arivis, HALO, Imaris) were considered. We focused on three key aspects: (a) the capacity to handle large file formats such as SVS, DICOM, and TIFF, ensuring compatibility with diverse datasets; (b) the flexibility in defining irregular ROIs, whether through automated extraction or manual delineation, encompassing square, circular, polygonal, and freehand shapes to accommodate varied research needs; and (c) the capability to classify single cells within selected ROIs on brightfield images, ranging from fully automated to semi-automated or manual approaches, requiring different levels of user involvement. Thanks to this work, a deeper understanding of the strengths and limitations of different software platforms emerges, facilitating informed decision making for researchers looking for a tool to analyse histological brightfield images. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

Figure 1
Main GUI of different open-source single-cell analysis tools. From top to bottom and left to right: CellProfiler, Cytomine, Digital Slide Archive, Icy, ImageJ/Fiji, and QuPath. Full article ">Figure 2
Main GUI of different freely available single-cell analysis tools. From top to bottom and left to right: Aperio ImageScope, NIS-Elements Viewer, Sedeen, and SlideViewer. Full article ">Figure 3
Main GUI of different commercial single-cell analysis tools. From top to bottom and left to right: Amira, Arivis, HALO, and Imaris. Full article ">Figure 4
Brightfield dataset used for comparing the different single-cell analysis tools. (a) Representative brightfield RGB image; (b) binary mask representing an irregular freehand-defined ROI; (c) original high-resolution WSI file with the position of the extracted image highlighted in red; (d) haematoxylin-channel image; (e) DAB-channel image; and (f) AP-red-channel image. Full article ">Figure 5
Overview of the segmentations obtained using the different tools. From top to bottom and left to right: Amira, Arivis, HALO, ImageJ/Fiji, NIS-Elements, and QuPath. Full article ">

12 pages, 932 KiB

Open AccessArticle

Enhancing Computational Efficiency of Network Reliability with a New Prime Shortest Path Algorithm

by Wei-Chang Yeh, Yunzhi Jiang and Chia-Ling Huang

Technologies 2025, 13(3), 109; https://doi.org/10.3390/technologies13030109 - 7 Mar 2025

Viewed by 77

Abstract

To address the increasing demands of modern networks, evaluating computational efficiency of modified network reliability is essential, with minimal paths (MPs) serving as a critical factor. However, traditional approaches to assessing computational efficiency of network reliability often struggle with challenges such as duplicate [...] Read more.

To address the increasing demands of modern networks, evaluating computational efficiency of modified network reliability is essential, with minimal paths (MPs) serving as a critical factor. However, traditional approaches to assessing computational efficiency of network reliability often struggle with challenges such as duplicate MPs and sub-path identification, resulting in exponential computational time. In this study, we present a novel algorithm based on the Prime Shortest Path (PSP) approach, which efficiently resolves these challenges by self-detecting and eliminating duplication in polynomial time. This marks a significant improvement over existing methods. The algorithm’s correctness is rigorously validated, and its superior performance is confirmed through a detailed time complexity analysis and comparisons with the leading state-of-the-art algorithms. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

20 pages, 6134 KiB

Open AccessArticle

A Hardware-in-the-Loop Simulation Platform for a High-Speed Maglev Positioning and Speed Measurement System

by Linzi Yin, Cong Luo, Ling Liu, Junfeng Cui, Zhiming Liu and Guoying Sun

Technologies 2025, 13(3), 108; https://doi.org/10.3390/technologies13030108 - 6 Mar 2025

Viewed by 145

Abstract

In order to solve the testing and verification problems at the early development stage of a high-speed Maglev positioning and speed measurement system (MPSS), a hardware-in-the-loop (HIL) simulation platform is presented, which includes induction loops, transmitting antennas, a power driver unit, a simulator [...] Read more.

In order to solve the testing and verification problems at the early development stage of a high-speed Maglev positioning and speed measurement system (MPSS), a hardware-in-the-loop (HIL) simulation platform is presented, which includes induction loops, transmitting antennas, a power driver unit, a simulator based on a field-programmable gate array (FPGA), a host computer, etc. This HIL simulation platform simulates the operation of a high-speed Maglev train and generates the related loop-induced signals to test the performance of a real ground signal processing unit (GSPU). Furthermore, an absolute position detection method based on Gray-coded loops is proposed to identify which Gray-coded period the train is in. A relative position detection method based on height compensation is also proposed to calculate the exact position of the train in a Gray-coded period. The experimental results show that the positioning error is only 2.58 mm, and the speed error is 6.34 km/h even in the 600 km/h condition. The proposed HIL platform also effectively simulates the three kinds of operation modes of high-speed Maglev trains, which verifies the effectiveness and practicality of the HIL simulation strategy. This provides favorable conditions for the development and early validation of high-speed MPSS. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

15 pages, 337 KiB

Open AccessArticle

Harnessing Metacognition for Safe and Responsible AI

by Peter B. Walker, Jonathan J. Haase, Melissa L. Mehalick, Christopher T. Steele, Dale W. Russell and Ian N. Davidson

Technologies 2025, 13(3), 107; https://doi.org/10.3390/technologies13030107 - 6 Mar 2025

Viewed by 103

Abstract

The rapid advancement of artificial intelligence (AI) technologies has transformed various sectors, significantly enhancing processes and augmenting human capabilities. However, these advancements have also introduced critical concerns related to the safety, ethics, and responsibility of AI systems. To address these challenges, the principles [...] Read more.

The rapid advancement of artificial intelligence (AI) technologies has transformed various sectors, significantly enhancing processes and augmenting human capabilities. However, these advancements have also introduced critical concerns related to the safety, ethics, and responsibility of AI systems. To address these challenges, the principles of the robustness, interpretability, controllability, and ethical alignment framework are essential. This paper explores the integration of metacognition—defined as “thinking about thinking”—into AI systems as a promising approach to meeting these requirements. Metacognition enables AI systems to monitor, control, and regulate the system’s cognitive processes, thereby enhancing their ability to self-assess, correct errors, and adapt to changing environments. By embedding metacognitive processes within AI, this paper proposes a framework that enhances the transparency, accountability, and adaptability of AI systems, fostering trust and mitigating risks associated with autonomous decision-making. Additionally, the paper examines the current state of AI safety and responsibility, discusses the applicability of metacognition to AI, and outlines a mathematical framework for incorporating metacognitive strategies into active learning processes. The findings aim to contribute to the development of safe, responsible, and ethically aligned AI systems. Full article

(This article belongs to the Special Issue Artificial Intelligence and Smart Information Systems: Trends and Innovations)

24 pages, 20905 KiB

Open AccessArticle

A Realistic Breast Phantom for Investigating the Features of the Microwave Radiometry Method Using Mathematical and Physical Modelling

by Maxim V. Polyakov and Danila S. Sirotin

Technologies 2025, 13(3), 106; https://doi.org/10.3390/technologies13030106 - 6 Mar 2025

Viewed by 235

Abstract

This article presents the development of an anatomical breast phantom for investigating the capabilities of microwave radiometry in assessing thermal processes in biological tissues. The phantom accounts for the heterogeneous tissue structure and haemodynamics, enabling realistic heat transfer modelling. Numerical simulation software was [...] Read more.

This article presents the development of an anatomical breast phantom for investigating the capabilities of microwave radiometry in assessing thermal processes in biological tissues. The phantom accounts for the heterogeneous tissue structure and haemodynamics, enabling realistic heat transfer modelling. Numerical simulation software was developed, accurately reproducing experimental results and allowing the study of thermal anomalies. Experimental validation demonstrated that the temperature in the subcutaneous layer differed on average by 0.3 °C from deeper tissues, confirming the method’s effectiveness. The presence of a tumour in the model resulted in a local temperature increase of up to 0.77 °C, highlighting the sensitivity of microwave radiometry to tumour-induced thermal anomalies. These findings contribute to enhancing non-invasive techniques for early breast disease detection. Full article

► Show Figures

Figure 1

34 pages, 10596 KiB

Open AccessArticle

Scalable Container-Based Time Synchronization for Smart Grid Data Center Networks

by Kennedy Chinedu Okafor, Wisdom Onyema Okafor, Omowunmi Mary Longe, Ikechukwu Ignatius Ayogu, Kelvin Anoh and Bamidele Adebisi

Technologies 2025, 13(3), 105; https://doi.org/10.3390/technologies13030105 - 5 Mar 2025

Viewed by 369

Abstract

The integration of edge-to-cloud infrastructures in smart grid (SG) data center networks requires scalable, efficient, and secure architecture. Traditional server-based SG data center architectures face high computational loads and delays. To address this problem, a lightweight data center network (DCN) with low-cost, and fast-converging [...] Read more.

The integration of edge-to-cloud infrastructures in smart grid (SG) data center networks requires scalable, efficient, and secure architecture. Traditional server-based SG data center architectures face high computational loads and delays. To address this problem, a lightweight data center network (DCN) with low-cost, and fast-converging optimization is required. This paper introduces a container-based time synchronization model (CTSM) within a spine–leaf virtual private cloud (SL-VPC), deployed via AWS CloudFormation stack as a practical use case. The CTSM optimizes resource utilization, security, and traffic management while reducing computational overhead. The model was benchmarked against five DCN topologies—DCell, Mesh, Skywalk, Dahu, and Ficonn—using Mininet simulations and a software-defined CloudFormation stack on an Amazon EC2 HPC testbed under realistic SG traffic patterns. The results show that CTSM achieved near-100% reliability, with the highest received energy data (29.87%), lowest packetization delay (13.11%), and highest traffic availability (70.85%). Stateless container engines improved resource allocation, reducing administrative overhead and enhancing grid stability. Software-defined Network (SDN)-driven adaptive routing and load balancing further optimized performance under dynamic demand conditions. These findings position CTSM-SL-VPC as a secure, scalable, and efficient solution for next-generation smart grid automation. Full article

► Show Figures

Figure 1

20 pages, 903 KiB

Open AccessArticle

A Hybrid Solar–Thermoelectric System Incorporating Molten Salt for Sustainable Energy Storage Solutions

by Mahmoud Z. Mistarihi, Ghazi M. Magableh and Saba M. Abu Dalu

Technologies 2025, 13(3), 104; https://doi.org/10.3390/technologies13030104 - 5 Mar 2025

Viewed by 240

Abstract

Green sustainable energy, especially renewable energy, is gaining huge popularity and is considered a vital energy in addressing energy conservation and global climate change. One of the most significant renewable energy sources in the UAE is solar energy, due to the country’s high [...] Read more.

Green sustainable energy, especially renewable energy, is gaining huge popularity and is considered a vital energy in addressing energy conservation and global climate change. One of the most significant renewable energy sources in the UAE is solar energy, due to the country’s high solar radiation levels. This paper focuses on advanced technology that integrates parabolic trough mirrors, molten salt storage, and thermoelectric generators (TEGs) to provide a reliable and effective solar system in the UAE. Furthermore, the new system can be manufactured in different sizes suitable for consumption whether in ordinary houses or commercial establishments and businesses. The proposed design theoretically achieves the target electrical energy of 2.067 kWh/day with 90% thermal efficiency, 90.2% optical efficiency, and 8% TEG efficiency that can be elevated to higher values reaching 149% using the liquid-saturated porous medium, ensuring the operation of the system throughout the day. This makes it a suitable solar system in off-grid areas. Moreover, this system is a cost-effective, carbon-free, and day-and-night energy source that can be dispatched on the electric grid like any fossil fuel plant under the proposed method, with less maintenance, thus contributing to the UAE’s renewable energy strategy. Full article

(This article belongs to the Section Environmental Technology)

► Show Figures

Figure 1

40 pages, 3792 KiB

Open AccessReview

Recent Development of Corrosion Inhibitors: Types, Mechanisms, Electrochemical Behavior, Efficiency, and Environmental Impact

by Denisa-Ioana (Gheorghe) Răuță, Ecaterina Matei and Sorin-Marius Avramescu

Technologies 2025, 13(3), 103; https://doi.org/10.3390/technologies13030103 - 5 Mar 2025

Viewed by 319

Abstract

This review examines recent advances in corrosion inhibitor technologies, with a focus on sustainable and environmentally friendly solutions that address both industrial efficiency and environmental safety. Corrosion is a ubiquitous problem, contributing to massive economic losses globally, with costs estimated between 1 and [...] Read more.

This review examines recent advances in corrosion inhibitor technologies, with a focus on sustainable and environmentally friendly solutions that address both industrial efficiency and environmental safety. Corrosion is a ubiquitous problem, contributing to massive economic losses globally, with costs estimated between 1 and 5% of GDP in different countries. Traditional inorganic corrosion inhibitors, while effective, are often based on toxic compounds, necessitating the development of more environmentally friendly and non-toxic alternatives. The present work highlights innovative eco-friendly corrosion inhibitors derived from natural sources, including plant extracts and oils, biopolymers, etc., being biodegradable substances that provide effective corrosion resistance with minimal environmental impact. In addition, this review explores organic–inorganic hybrid inhibitors and nanotechnology-enhanced coatings that demonstrate improved efficiency, durability, and adaptability across industries. Key considerations, such as application techniques, mechanisms of action, and the impact of environmental factors on inhibitor performance, are discussed. This comprehensive presentation aims to contribute to updating the data on the development of advanced corrosion inhibitors capable of meeting the requirements of modern industries while promoting sustainable and safe practices in corrosion management. Full article

(This article belongs to the Section Environmental Technology)

► Show Figures

Figure 1

76 pages, 8958 KiB

Open AccessArticle

Robust Intrusion Detection System Using an Improved Hybrid Deep Learning Model for Binary and Multi-Class Classification in IoT Networks

by Hesham Kamal and Maggie Mashaly

Technologies 2025, 13(3), 102; https://doi.org/10.3390/technologies13030102 - 4 Mar 2025

Viewed by 301

Abstract

The rapid expansion of internet of things (IoT) applications has significantly boosted productivity and streamlined daily activities. However, this widespread adoption has also introduced considerable security challenges, making IoT environments vulnerable to large-scale botnet attacks. These attacks have often succeeded in achieving their [...] Read more.

The rapid expansion of internet of things (IoT) applications has significantly boosted productivity and streamlined daily activities. However, this widespread adoption has also introduced considerable security challenges, making IoT environments vulnerable to large-scale botnet attacks. These attacks have often succeeded in achieving their malicious goals, highlighting the urgent need for robust detection strategies to secure IoT networks. To overcome these obstacles, this research presents an innovative anomaly-driven intrusion detection approach specifically tailored for IoT networks. The proposed model employs an advanced hybrid architecture that seamlessly integrates convolutional neural networks (CNN) with multilayer perceptron (MLP), enabling precise detection and classification of both binary and multi-class IoT network traffic. The CNN component is responsible for extracting and enhancing features from network traffic data and preparing these features for effective classification by the MLP, which handles the final classification task. To further manage class imbalance, the model incorporates the enhanced hybrid adaptive synthetic sampling-synthetic minority oversampling technique (ADASYN-SMOTE) for binary classification, advanced ADASYN for multiclass classification, and employs edited nearest neighbors (ENN) alongside class weights. The CNN-MLP architecture is meticulously crafted to minimize erroneous classifications, enhance instantaneous threat detection, and precisely recognize previously unseen cyber intrusions. The model’s effectiveness was rigorously tested using the IoT-23 and NF-BoT-IoT-v2 datasets. On the IoT-23 dataset, the model achieved 99.94% accuracy in two-stage binary classification, 99.99% accuracy in multiclass classification excluding the normal class, and 99.91% accuracy in single-phase multiclass classification including the normal class. Utilizing the NF-BoT-IoT-v2 dataset, the model attained an exceptional 99.96% accuracy in the dual-phase binary classification paradigm, 98.02% accuracy in multiclass classification excluding the normal class, and 98.11% accuracy in single-phase multiclass classification including the normal class. The results demonstrate that our model consistently delivers high levels of accuracy, precision, recall, and F1 score across both binary and multiclass classifications, establishing it as a robust solution for securing IoT networks. Full article

► Show Figures

Figure 1

20 pages, 2485 KiB

Open AccessArticle

Hand Dexterity Evaluation Grounded on Cursor Trajectory Investigation in Children with ADHD Using a Mouse and a Joystick

by Alexandros Pino, Nikolaos Papatheodorou, Georgios Kouroupetroglou, Panagiotis-Alexios Giannopoulos, Gerasimos Makris and Charalambos Papageorgiou

Technologies 2025, 13(3), 99; https://doi.org/10.3390/technologies13030099 - 3 Mar 2025

Viewed by 278

Abstract

This study investigates disparities in upper limb motor skills between children with and without Attention Deficit Hyperactivity Disorder (ADHD), employing one-dimensional (1D) and two-dimensional (2D) point-and-click experiments using a mouse and a joystick and introducing one new metric for mouse cursor trajectory analysis. [...] Read more.

This study investigates disparities in upper limb motor skills between children with and without Attention Deficit Hyperactivity Disorder (ADHD), employing one-dimensional (1D) and two-dimensional (2D) point-and-click experiments using a mouse and a joystick and introducing one new metric for mouse cursor trajectory analysis. The participant pool comprised 46 children with combined type ADHD and an equivalent number of children without ADHD. The Input Device Evaluation Application (IDEA) system monitored the mouse pointer’s trajectory. Ten trajectory parameters were computed, including Index of Difficulty, Movement Time, Throughput, Missed Clicks, Target Re-Entry, Task Axis Crossing, Movement Direction Change, Movement Variability, Movement Error, Movement Offset, and Sample Entropy. The 2D joystick experiment trajectory parameters analysis conducted using a hierarchical logistic regression model achieved a 78% success rate in identifying children with ADHD. This research sheds light on the motor skill differences associated with ADHD in the context of computer-based tasks, providing valuable insights into potential diagnostic applications and intervention strategies and introducing one new metric makes for a deeper cursor trajectory analysis. Full article

(This article belongs to the Section Assistive Technologies)

► Show Figures

Figure 1

28 pages, 7320 KiB

Open AccessArticle

Technology for Improving the Accuracy of Predicting the Position and Speed of Human Movement Based on Machine Learning Models

by Artem Obukhov, Denis Dedov, Andrey Volkov and Maksim Rybachok

Technologies 2025, 13(3), 101; https://doi.org/10.3390/technologies13030101 - 3 Mar 2025

Viewed by 386

Abstract

The solution to the problem of insufficient accuracy in determining the position and speed of human movement during interaction with a treadmill-based training complex is considered. Control command generation based on the training complex user’s actions may be performed with a delay, may [...] Read more.

The solution to the problem of insufficient accuracy in determining the position and speed of human movement during interaction with a treadmill-based training complex is considered. Control command generation based on the training complex user’s actions may be performed with a delay, may not take into account the specificity of movements, or be inaccurate due to the error of the initial data. The article introduces a technology for improving the accuracy of predicting a person’s position and speed on a running platform using machine learning and computer vision methods. The proposed technology includes analysing and processing data from the tracking system, developing machine learning models to improve the quality of the raw data, predicting the position and speed of human movement, and implementing and integrating neural network methods into the running platform control system. Experimental results demonstrate that the decision tree (DT) model provides better accuracy and performance in solving the problem of positioning key points of a human model in complex conditions with overlapping limbs. For speed prediction, the linear regression (LR) model showed the best results when the analysed window length was 10 frames. Prediction of the person’s position (based on 10 previous frames) is performed using the DT model, which is optimal in terms of accuracy and computation time relative to other options. The comparison of the control methods of the running platform based on machine learning models showed the advantage of the combined method (linear control function combined with the speed prediction model), which provides an average absolute error value of 0.116 m/s. The results of the research confirmed the achievement of the primary objective (increasing the accuracy of human position and speed prediction), making the proposed technology promising for application in human-machine systems. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

20 pages, 2678 KiB

Open AccessArticle

Low-Temperature Slow Pyrolysis: Exploring Biomass-Specific Biochar Characteristics and Potential for Soil Applications

by Matheus Antonio da Silva, Adibe Luiz Abdalla Filho, Ruan Carnier, Juliana de Oliveira Santos Marcatto, Marcelo Saldanha, Aline Renee Coscione, Thaís Alves de Carvalho, Gabriel Rodrigo Merlotto and Cristiano Alberto de Andrade

Technologies 2025, 13(3), 100; https://doi.org/10.3390/technologies13030100 - 3 Mar 2025

Viewed by 339

Abstract

The pyrolysis process of residues has emerged as a sustainable method for managing organic waste, producing biochars that offer significant benefits for agriculture and the environment. These benefits depend on the properties of the raw biomass and the pyrolysis conditions, such as washing [...] Read more.

The pyrolysis process of residues has emerged as a sustainable method for managing organic waste, producing biochars that offer significant benefits for agriculture and the environment. These benefits depend on the properties of the raw biomass and the pyrolysis conditions, such as washing and drying. This study investigated biochar production through slow pyrolysis at 300 °C, using eight biomass types, four being plant residues (PBR)—sugarcane bagasse, filter cake, sawdust, and stranded algae—and four non-plant-based residues (NPBR)—poultry litter, sheep manure, layer chicken manure, and sewage sludge. The physicochemical properties assessed included yield, carbon (C) and nitrogen (N) content, electrical conductivity, pH, macro- and micronutrients, and potentially toxic metals. Pyrolysis generally increased pH and concentrated C, N, phosphorus (P), and other nutrients while reducing electrical conductivity, C/N ratio, potassium (K), and sulfur (S) contents. The increases in the pH of the biochars in relation to the respective biomasses were between 0.3 and 1.9, with the greatest differences observed for the NPBR biochars. Biochars from sugarcane bagasse and sawdust exhibited high C content (74.57–77.67%), highlighting their potential use for C sequestration. Filter cake biochar excelled in P (14.28 g kg⁻¹) and micronutrients, while algae biochar showed elevated N, calcium (Ca), and boron (B) levels. NPBR biochars were rich in N (2.28–3.67%) and P (20.7–43.4 g kg⁻¹), making them ideal fertilizers. Although sewage sludge biochar contained higher levels of potentially toxic metals, these remained within regulatory limits. This research highlights variations in the composition of biochars depending on the characteristics of the original biomass and the pyrolysis process, to contribute to the production of customized biochars for the purposes of their application in the soil. Biochars derived from exclusively plant biomasses showed important aspects related to the recovery of carbon from biomass and can be preferred as biochar used to sequester carbon in the soil. On the other hand, biochars obtained from residues with some animal contributions are more enriched in nutrients and should be directed to the management of soil fertility. Full article

(This article belongs to the Special Issue Recent Advances in Applied Activated Carbon Research)

► Show Figures

Figure 1

Figure 1
Total C results in the biomasses and their respective biochars, before and after washing. (A) total C content and (B) mass yield and C recovery. The vertical bars represent the 5% confidence interval. Full article ">Figure 2
Organic C results in the biomasses and their respective biochars. (A) organic C content and (B) recovery percentage of organic C. The vertical bars represent the 5% confidence interval. Full article ">Figure 3
Total N results in the biomasses and their respective biochars, before and after washing. (A) total N content and (B) total N recoverey. The vertical bars represent the 5% confidence interval. Full article ">Figure 4
Extractable N results in the biomasses and their respective biochars. (A) extractable N content and (B) extractable N recovery. The vertical bars represent the 5% confidence interval. Full article ">Figure 5
C/N ratio for biomasses and the respective biochars, (A) from PBR and (B) from NPBR. Full article ">Figure 6
S and K results for all biochar and its respective biomasses. (A) Total S content, (B) Recovery of S, (C) Total K content and (D) Recovery of K. The vertical bars represent the 5% confidence interval. Full article ">

25 pages, 42227 KiB

Open AccessArticle

“The Foot Can Do It”: Controlling the “Persistence” Prosthetic Arm Using the “Infinity-2” Foot Controller

by Peter L. Bishay, Gerbert Funes Alfaro, Ian Sherrill, Isaiah Reoyo, Elihu McMahon, Camron Carter, Cristian Valdez, Naweeth M. Riyaz, Sara Ali, Adrian Lima, Abel Nieto and Jared Tirone

Technologies 2025, 13(3), 98; https://doi.org/10.3390/technologies13030098 - 1 Mar 2025

Viewed by 483

Abstract

The “Infinity” foot controller for controlling prosthetic arms has been improved in this paper in several ways, including a foot sleeve that enables barefoot use, an improved sensor-controller unit design, and a more intuitive control scheme that allows gradual control of finger actuation. [...] Read more.

The “Infinity” foot controller for controlling prosthetic arms has been improved in this paper in several ways, including a foot sleeve that enables barefoot use, an improved sensor-controller unit design, and a more intuitive control scheme that allows gradual control of finger actuation. Furthermore, the “Persistence Arm”, a novel transradial prosthetic arm prototype, is introduced. This below-the-elbow arm has a direct-drive wrist actuation system, a thumb design with two degrees of freedom, and carbon fiber tendons for actuating the four forefingers. The manufactured prototype arm and foot controller underwent various tests to verify their efficacy. Wireless transmission speed tests showed that the maximum time delay is less than 165 ms, giving almost instantaneous response from the arm to any user’s foot control signal. Gripping tests quantified the grip and pulling forces of the arm prototype as 2.8 and 12.7 kg, respectively. The arm successfully gripped various household items of different shapes, weights, and sizes. These results highlight the potential of foot control as an alternative prosthetic arm control method and the possibility of new 3D-printed prosthetic arm designs to replace costly prostheses in the market, which could potentially reduce the high rejection rates of upper limb prostheses. Full article

(This article belongs to the Special Issue Assistive Technologies in Care and Rehabilitation: Research, Developments, and International Initiatives)

► Show Figures

Graphical abstract

21 pages, 12426 KiB

Open AccessArticle

Scientific Molding and Adaptive Process Quality Control with External Sensors for Injection Molding Process

by Chen-Hsiang Chang, Chien-Hung Wen, Ren-Ho Tseng, Chieh-Hsun Tsai, Yu-Hao Chen, Sheng-Jye Hwang and Hsin-Shu Peng

Technologies 2025, 13(3), 97; https://doi.org/10.3390/technologies13030097 - 1 Mar 2025

Viewed by 380

Abstract

This study established a real-time measurement system to monitor the melt quality in an injection molding process using a pressure sensor installed on the nozzle and a strain gauge installed on the tie bar. Based on the sensing curves from these two external [...] Read more.

This study established a real-time measurement system to monitor the melt quality in an injection molding process using a pressure sensor installed on the nozzle and a strain gauge installed on the tie bar. Based on the sensing curves from these two external sensors, the characteristic values of nozzle pressure and clamping force were used to optimize parameters. This study defined product weight as a quality indicator and developed a scientific molding parameter setup process. The optimization sequence of parameters is injection speed, V/P switchover point, packing pressure, packing time, and clamping force. Finally, an adaptive process control system was established based on the online quality characteristic values to maintain product quality consistency. Continuous production experiments were conducted at two sites to verify the system’s effectiveness. The results revealed that the optimized process parameters can ensure product weight stability during long-term production. Furthermore, using the adaptive process control system further enhanced product weight stability at both sites, reducing the standard deviation of product weight to 0.0289 g and 0.0148 g, and the coefficient of variation to 0.065% and 0.035%, respectively. Full article

(This article belongs to the Section Manufacturing Technology)

► Show Figures

Graphical abstract

12 pages, 375 KiB

Open AccessProtocol

Training Cognitive Functions Using DUAL-REHAB, a New Dual-Task Application in MCI and SMC: A Study Protocol of a Randomized Control Trial

by Elisa Pedroli, Francesca Bruni, Valentina Mancuso, Silvia Cavedoni, Francesco Bigotto, Jonathan Panigada, Monica Rossi, Lorenzo Boilini, Karine Goulene, Marco Stramba-Badiale and Silvia Serino

Technologies 2025, 13(3), 96; https://doi.org/10.3390/technologies13030096 - 1 Mar 2025

Viewed by 343

Abstract

Background: Current research on Alzheimer’s Disease has progressively focused on Mild Cognitive Impairment (MCI) as a pre-dementia state, as well as on Subjective Memory Complaint (SMC), as a potential early indicator of cognitive change. Consequently, timely interventions to prevent cognitive decline are essential [...] Read more.

Background: Current research on Alzheimer’s Disease has progressively focused on Mild Cognitive Impairment (MCI) as a pre-dementia state, as well as on Subjective Memory Complaint (SMC), as a potential early indicator of cognitive change. Consequently, timely interventions to prevent cognitive decline are essential and are most effective when combined with motor training. Nevertheless, motor-cognitive dual-task training often employs non-ecological tasks and is confined to clinical contexts lacking generalizability to daily life. The integration of 360° media could overcome these limitations. Therefore, the aim of the current work is twofold: (a) to present a dual-task training using 360° technology for its interactivity, versatility, and ecological validity, and (b) to propose a protocol to test its efficacy through a randomized clinical trial. Methods: This study will recruit 90 older adults (MCI and SMC). Participants will follow two phases of training: in-hospital rehabilitation and at-home rehabilitation. The experimental design will follow a 2 × 3 × 2 structure with 3 factors: type of treatment (360° training vs. traditional rehabilitation), time (baseline, post in-hospital training, and post at-home training), and group (SMC vs. MCI). Results: The expected outcome is an improvement in cognitive and motor functioning after the experimental training. Conclusion: This study will advance the literature on non-pharmacological interventions and innovative technological tools for cognitive trainings in the early stages of cognitive decline. Full article

(This article belongs to the Special Issue Innovations in Design, Development and Evaluation of Assistive Technologies)

► Show Figures

Figure 1

18 pages, 11708 KiB

Open AccessArticle

The Transition to an Eco-Friendly City as a First Step Toward Climate Neutrality with Green Hydrogen

by Lăzăroiu Gheorghe, Mihăescu Lucian, Stoica Dorel and Năstasă (Băcăran) Florentina-Cătălina

Technologies 2025, 13(3), 95; https://doi.org/10.3390/technologies13030095 - 1 Mar 2025

Viewed by 340

Abstract

A city of the future will need to be eco-friendly while meeting general social and economic requirements. Hydrogen-based technologies provide solutions for initially limiting CO₂ emissions, with prospects indicating complete decarbonization in the future. Cities will need to adopt and integrate these [...] Read more.

A city of the future will need to be eco-friendly while meeting general social and economic requirements. Hydrogen-based technologies provide solutions for initially limiting CO₂ emissions, with prospects indicating complete decarbonization in the future. Cities will need to adopt and integrate these technologies to avoid a gap between the development of hydrogen production and its urban application. Achievable results are analyzed by injecting hydrogen into the urban methane gas network, initially in small proportions, but gradually increasing over time. This paper also presents a numerical application pertaining to the city of Bucharest, Romania—a metropolis with a population of 2.1 million inhabitants. Although the use of fuel cells is less advantageous for urban transport compared to electric battery-based solutions, the heat generated by hydrogen-based technologies, such as fuel cells, can be efficiently utilized for residential heating. However, storage solutions are required for residential consumption, separate from that of urban transport, along with advancements in electric transport using existing batteries, which necessitate a detailed economic assessment. For electricity generation, including cogeneration, gas turbines have proven to be the most suitable solution. Based on the analyzed data, the paper synthesizes the opportunities offered by hydrogen-based technologies for a city of the future. Full article

(This article belongs to the Section Environmental Technology)

► Show Figures

Figure 1

45 pages, 5675 KiB

Open AccessReview

A Comprehensive Review of Quality Control and Reliability Research in Micro–Nano Technology

by Nowshin Sharmile, Risat Rimi Chowdhury and Salil Desai

Technologies 2025, 13(3), 94; https://doi.org/10.3390/technologies13030094 - 1 Mar 2025

Viewed by 457

Abstract

This paper presents a comprehensive review of quality control (QC) and reliability research in micro–nano technology, which is vital for advancing microelectronics, biomedical engineering, and manufacturing. Micro- and nanotechnologies operate at different scales, yet both require precise control to ensure the performance and [...] Read more.

This paper presents a comprehensive review of quality control (QC) and reliability research in micro–nano technology, which is vital for advancing microelectronics, biomedical engineering, and manufacturing. Micro- and nanotechnologies operate at different scales, yet both require precise control to ensure the performance and durability of small-scale systems. This review synthesizes key quality control methodologies, including statistical quality control methods, machine learning and AI-driven methods, and advanced techniques emphasizing their relevance to nanotechnology applications. The paper also discusses the application of micro/nanotechnology in quality control in other technological areas. The discussion extends to the unique reliability challenges posed by micro–nano systems, such as failure modes related to stiction, material fatigue, and environmental factors. Advanced reliability testing and modeling approaches are highlighted for their effectiveness in predicting performance and mitigating risks. Additionally, the paper explores the integration of emerging technologies to enhance and improve reliability in micro–nano manufacturing. By examining both established and novel techniques, this review underscores the evolving nature of quality control and reliability research in the field. It identifies key areas for future investigation, particularly in the adaptation of these methods to the increasing complexity of micro–nano systems. The paper concludes by proposing research directions that can further optimize quality control and reliability to ensure the continued advancement and industrial application of micro–nano technologies. Full article

(This article belongs to the Section Innovations in Materials Processing)

► Show Figures

Figure 1

31 pages, 4379 KiB

Open AccessSystematic Review

A Systematic Literature Review of the Latest Advancements in XAI

by Zaid M. Altukhi, Sojen Pradhan and Nasser Aljohani

Technologies 2025, 13(3), 93; https://doi.org/10.3390/technologies13030093 - 1 Mar 2025

Viewed by 425

Abstract

This systematic review details recent advancements in the field of Explainable Artificial Intelligence (XAI) from 2014 to 2024. XAI utilises a wide range of frameworks, techniques, and methods used to interpret machine learning (ML) black-box models. We aim to understand the technical advancements [...] Read more.

This systematic review details recent advancements in the field of Explainable Artificial Intelligence (XAI) from 2014 to 2024. XAI utilises a wide range of frameworks, techniques, and methods used to interpret machine learning (ML) black-box models. We aim to understand the technical advancements in the field and future directions. We followed the PRISMA methodology and selected 30 relevant publications from three main databases: IEEE Xplore, ACM, and ScienceDirect. Through comprehensive thematic analysis, we categorised the research into three main topics: ‘model developments’, ‘evaluation metrics and methods’, and ‘user-centred and XAI system design’. Our results uncover ‘What’, ‘How’, and ‘Why’ these advancements were developed. We found that 13 papers focused on model developments, 8 studies focused on the XAI evaluation metrics, and 12 papers focused on user-centred and XAI system design. Moreover, it was found that these advancements aimed to bridge the gap between technical model outputs and user understanding. Full article

► Show Figures

Figure 1

26 pages, 3719 KiB

Open AccessArticle

Design of Multi-Sourced MIMO Multiband Hybrid Wireless RF-Perovskite Photovoltaic Energy Harvesting Subsystems for IoTs Applications in Smart Cities

by Fanuel Elias, Sunday Ekpo, Stephen Alabi, Mfonobong Uko, Sunday Enahoro, Muhammad Ijaz, Helen Ji, Rahul Unnikrishnan and Nurudeen Olasunkanmi

Technologies 2025, 13(3), 92; https://doi.org/10.3390/technologies13030092 - 1 Mar 2025

Viewed by 492

Abstract

Energy harvesting technology allows Internet of Things (IoT) devices to be powered continuously without needing battery charging or replacement. In addressing existing and emerging massive IoT energy supply challenges, this paper presents the design of multi-sourced multiple input and multiple output (MIMO) multiband [...] Read more.

Energy harvesting technology allows Internet of Things (IoT) devices to be powered continuously without needing battery charging or replacement. In addressing existing and emerging massive IoT energy supply challenges, this paper presents the design of multi-sourced multiple input and multiple output (MIMO) multiband hybrid wireless RF-perovskite photovoltaic energy harvesting subsystems for IoT application. The research findings evaluate the efficiency and power output of different RF configurations (1 to 16 antennas) within MIMO RF subsystems. A Delon quadruple rectifier in the RF energy harvesting system demonstrates a system-level power conversion efficiency of 51%. The research also explores the I-V and P-V characteristics of the adopted perovskite tandem cell. This results in an impressive array capable of producing 6.4 V and generating a maximum power of 650 mW. For the first time, the combined mathematical modelling of the system architecture is presented. The achieved efficiency of the combined system is 90% (for 8 MIMO) and 98% (for 16 MIMO) at 0 dBm input RF power. This novel study holds great promise for next-generation 5G/6G smart IoT passive electronics. Additionally, it establishes the hybrid RF-perovskite energy harvester as a promising, compact, and eco-friendly solution for efficiently powering IoT devices in smart cities. This work contributes to the development of sustainable, scalable, and smart energy solutions for IoT integration into smart city infrastructures. Full article

► Show Figures

Graphical abstract

18 pages, 5532 KiB

Open AccessArticle

Field Data Retrieval for Electric Vehicles and Estimating Equivalent Circuit Model Parameters via Particle Swarm Optimization

by Syed Adil Sardar, Shahzad Iqbal, Jeongju Park, Sekyung Han and Woo Young Kim

Technologies 2025, 13(3), 91; https://doi.org/10.3390/technologies13030091 - 1 Mar 2025

Viewed by 421

Abstract

Data retrieval techniques are crucial for developing an effective battery management system for an electric vehicle to accurately assess the battery’s health and performance by monitoring operating conditions such as voltage, current, time, temperature, and state of charge. This paper proposes an efficient [...] Read more.

Data retrieval techniques are crucial for developing an effective battery management system for an electric vehicle to accurately assess the battery’s health and performance by monitoring operating conditions such as voltage, current, time, temperature, and state of charge. This paper proposes an efficient approach to retrieve real-world field data (voltage, current, and time) under running vehicle conditions. In the first step, noise is removed from the field data using a moving-average filter. Then, first- and second-order derivations are applied to the filtered data to determine specific data set conditions. After that, a new approach based on zero-crossing is implemented to retrieve the field data. A second-order Randle circuit (SORC) is utilized in this study to analyze the selected field data. Further, a particle swarm optimization algorithm is adapted to estimate the parameters of the SORC. Our experiments indicate that the relative errors of the equivalent circuit model (ECM) are less than 2% compared to the model voltage and real voltage, which is consistent with the stable parameters of ECM. Full article

► Show Figures

Graphical abstract

16 pages, 4441 KiB

Open AccessArticle

Electrospinning of Chitosan–Halloysite Nanotube Biohybrid Mats for Clobetasol Propionate Delivery

by Natallia V. Dubashynskaya, Valentina A. Petrova, Igor V. Kudryavtsev, Andrey S. Trulioff, Artem A. Rubinstein, Alexey S. Golovkin, Alexander I. Mishanin, Anton A. Murav’ev, Iosif V. Gofman, Daria N. Poshina and Yury A. Skorik

Technologies 2025, 13(3), 90; https://doi.org/10.3390/technologies13030090 - 21 Feb 2025

Viewed by 457

Abstract

The application of electrospinning technologies for the preparation of mats based on mucoadhesive polysaccharides, such as chitosan (CS), is an attractive strategy for the development of biopolymeric delivery systems for topical corticosteroids. In this work, an electrospinning technique is described for the preparation [...] Read more.

The application of electrospinning technologies for the preparation of mats based on mucoadhesive polysaccharides, such as chitosan (CS), is an attractive strategy for the development of biopolymeric delivery systems for topical corticosteroids. In this work, an electrospinning technique is described for the preparation of CS-based mats doped with halloysite nanotubes (HNTs) with modified release of clobetasol propionate (CP). The optimized composition of the electrospinning solution was determined: 2.4% solution of CS in 46% acetic acid with addition of PEO (10% of CS mass) and HNTs (5% of CS mass); CP was introduced as an ethanol solution at the rate of 2 mg CP per 1 g of the obtained nonwoven material. The process parameters (the electrospinning voltage of 50–65 kV, the rotation speed of the spinning electrode of 10 min⁻¹, and the distance between the electrodes of 24 cm) were also optimized. The developed technology allowed us to obtain homogeneous nanofiber mats with excellent mechanical properties and biphasic drug release patterns (66% of CP released within 0.5 h and 88% of CP released within 6 h). The obtained nanofiber mats maintained the anti-inflammatory activity of corticosteroid at the level of free CP and showed no cytotoxicity. Full article

(This article belongs to the Section Innovations in Materials Processing)

► Show Figures

Figure 1

20 pages, 4114 KiB

Open AccessArticle

Effect of Pore Characteristics of Biomass-Derived Activated Carbon for Automobile Canisters via Chemical Stabilization Method on Butane Adsorption Characteristics

by Dong-Sin Jo, Ju-Hwan Kim, Byung-Joo Kim and Hye-Min Lee

Technologies 2025, 13(3), 89; https://doi.org/10.3390/technologies13030089 - 21 Feb 2025

Viewed by 356

Abstract

In this study, kenaf-derived activated carbons (AK-AC) was prepared for automobile canisters via chemical stabilization and physical activation methods. The thermogravimetric analysis and differential thermogravimetry revealed a crystallite change in the kenaf with chemical stabilization. The AK-AC texture properties were studied using the [...] Read more.

In this study, kenaf-derived activated carbons (AK-AC) was prepared for automobile canisters via chemical stabilization and physical activation methods. The thermogravimetric analysis and differential thermogravimetry revealed a crystallite change in the kenaf with chemical stabilization. The AK-AC texture properties were studied using the Brunauer–Emmett–Teller, Dubinin–Radushkevitch, and non-local density functional theory equations, with N₂/77K isotherm adsorption–desorption curves. The AK-AC nanocrystallite characteristics were observed through X-ray diffraction and Raman spectroscopy. The AK-AC butane adsorption characteristics were analyzed via breakthrough curves and compared with those of commercial coconut-derived activated carbon (Coconut AC). As the activation time increased, the specific surface area and mesopore volume ratio of the AK-AC increased to 1080–1940 m²/g and 10.6–50.0%, respectively. The AK-AC also exhibited better mesoporous pore characteristics than the Coconut AC. The AK-AC butane adsorption capacity increased from 0.31 to 0.79 g/g. In particular, the AK-AC had an approximately 50% improved butane adsorption capacity compared to the Coconut AC. In addition, the butane adsorption characteristics of the AK-AC were determined using the mesopore volume, with a diameter of 3.0–4.0 nm. The results suggest that AK-AC may be proposed as an adsorbent to improve evaporative emissions from automotive canisters in the future. Full article

(This article belongs to the Special Issue Recent Advances in Applied Activated Carbon Research)

► Show Figures

Figure 1

Figure 1
Schematic of butane adsorption simulation equipment. Full article ">Figure 2
(a) TGA and (b) DTG curves of the stabilized kenaf samples under N2 atmosphere. Full article ">Figure 3
XRD patterns derived as a function of various steam activation conditions: (a) stabilized kenaf and (b) AK-AC. Full article ">Figure 4
Structural characteristics of the AK-AC as a function of various steam activation conditions: (a) structure parameters and (b) interplanar distance. Full article ">Figure 5
(a) Raman spectra of AK-AC as a function of various activation conditions. (b) Band parameters derived from raw spectra decompositions. Full article ">Figure 6
FE-SEM images for (a) as-received sample, (b) K-AP-20-H-9-1, (c) K-AP-20-H-9-2, (d) K-AP-20-H-9-3, (e) K-AP-20-H-9-4, (f) K-AP-20-H-9-6, (g) K-AP-20-H-9-8, and (h) K-AP-20-H-9-10. Full article ">Figure 7
N2/77K isotherm adsorption/desorption curves of the AK-AC as a function of various activation conditions: (a) CK-AC and (b) AK-AC. Full article ">Figure 8
PSD as a function of various steam activation conditions by the NLDFT method: (a) CK-AC and (b) AK-AC. Full article ">Figure 9
Breakthrough curves of the AK-AC as a function of various activation conditions: (a) breakthrough curve and (b) breakthrough point. Full article ">Figure 10
Correlation between the butane capacity and pore volume of the AK-AC. Full article ">

40 pages, 4296 KiB

Open AccessArticle

Comprehensive Analysis of Random Forest and XGBoost Performance with SMOTE, ADASYN, and GNUS Under Varying Imbalance Levels

by Mehdi Imani, Ali Beikmohammadi and Hamid Reza Arabnia

Technologies 2025, 13(3), 88; https://doi.org/10.3390/technologies13030088 - 20 Feb 2025

Viewed by 846

Abstract

This study examines the efficacy of Random Forest and XGBoost classifiers in conjunction with three upsampling techniques—SMOTE, ADASYN, and Gaussian noise upsampling (GNUS)—across datasets with varying class imbalance levels, ranging from moderate to extreme (15% to 1% churn rate). Employing metrics such as [...] Read more.

This study examines the efficacy of Random Forest and XGBoost classifiers in conjunction with three upsampling techniques—SMOTE, ADASYN, and Gaussian noise upsampling (GNUS)—across datasets with varying class imbalance levels, ranging from moderate to extreme (15% to 1% churn rate). Employing metrics such as F1 score, ROC AUC, PR AUC, Matthews Correlation Coefficient (MCC), and Cohen’s Kappa, this research provides a comprehensive evaluation of classifier performance under different imbalance scenarios, focusing on applications in the telecommunications domain. The findings highlight that tuned XGBoost paired with SMOTE (Tuned_XGB_SMOTE) consistently achieves the highest F1 score and robust performance across all imbalance levels. SMOTE emerged as the most effective upsampling method, particularly when used with XGBoost, whereas Random Forest performed poorly under severe imbalance. ADASYN showed moderate effectiveness with XGBoost but underperformed with Random Forest, and GNUS produced inconsistent results. This study underscores the impact of data imbalance, with MCC, Kappa, and F1 scores fluctuating significantly, whereas ROC AUC and PR AUC remained relatively stable. Moreover, rigorous statistical analyses employing the Friedman test and Nemenyi post hoc comparisons confirmed that the observed improvements in F1 score, PR-AUC, Kappa, and MCC were statistically significant (p < 0.05), with Tuned_XGB_SMOTE significantly outperforming Tuned_RF_GNUS. While differences in ROC-AUC were not significant, the consistency of these results across multiple performance metrics underscores the reliability of our framework, offering a statistically validated and attractive solution for model selection in imbalanced classification scenarios. Full article

(This article belongs to the Special Issue Data Science and Big Data in Biology, Physical Science and Engineering II)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Technologies, Volume 13, Issue 3 (March 2025) – 25 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI