Applied System Innovation

11 pages, 1297 KB

Open AccessArticle

Switch Cabinet Temperature Prediction Using a Fusion of CNN and LSTM Neural Networks

by Xu Sun, Yun Chen, Jiang Wei, Qi Liu, Hui Guo and Ruijian Cheng

Appl. Syst. Innov. 2025, 8(5), 157; https://doi.org/10.3390/asi8050157 - 21 Oct 2025

This study investigates the effectiveness of a hybrid convolutional neural network (CNN) and long short-term memory (LSTM) model for predicting the temperature of switchgear within electrical power systems. Given the critical importance of temperature monitoring for operational safety and stability, this research integrates [...] Read more.

This study investigates the effectiveness of a hybrid convolutional neural network (CNN) and long short-term memory (LSTM) model for predicting the temperature of switchgear within electrical power systems. Given the critical importance of temperature monitoring for operational safety and stability, this research integrates CNNs and LSTMs to leverage their respective strengths in spatial feature extraction and temporal data processing. Utilizing a dataset from 2020 comprising hourly data points along with comprehensive environmental and operational variables, the model aims to deliver precise temperature predictions. Initial results indicate a high level of accuracy, with the CNN-LSTM model achieving an

R^{2}

score of 0.95 and a mean absolute error of 0.12 °C, highlighting its significant potential to enhance the monitoring and management of safety in power systems. Full article

► Show Figures

Figure 1

54 pages, 5577 KB

Open AccessArticle

Innovative Method for Detecting Malware by Analysing API Request Sequences Based on a Hybrid Recurrent Neural Network for Applied Forensic Auditing

by Serhii Vladov, Victoria Vysotska, Vitalii Varlakhov, Mariia Nazarkevych, Serhii Bolvinov and Volodymyr Piadyshev

Appl. Syst. Innov. 2025, 8(5), 156; https://doi.org/10.3390/asi8050156 - 21 Oct 2025

Abstract

This article develops a method for detecting malware based on the multi-scale recurrent architecture (time-aware multi-scale LSTM) with salience gating, multi-headed attention, and a sequential statistical change detector (CUSUM) integration. The research aim is to create an algorithm capable of effectively detecting malicious [...] Read more.

This article develops a method for detecting malware based on the multi-scale recurrent architecture (time-aware multi-scale LSTM) with salience gating, multi-headed attention, and a sequential statistical change detector (CUSUM) integration. The research aim is to create an algorithm capable of effectively detecting malicious activities in behavioural data streams of executable files with minimal delay and ensuring interpretability of the results for subsequent use in forensic audit and cyber defence systems. To implement the task, deep learning methods (training LSTM models with dynamic consideration of time intervals and adaptive attention mechanisms) and sequence statistical analysis (CUSUM, Kulback–Leibler divergence, and Wasserstein distances), as well as regularisation approaches to improve the model stability and explainability, were used. Experimental evaluation demonstrates the proposed approaches’ high efficiency, with the neural network model achieving competitive indicators of accuracy, recall, and classification balance with a low level of false positives and an acceptable detection delay. Attention and salience profile analysis confirmed the possibility of interpreting signals and early detection of abnormal events, which reduces the experts’ workload and reduces the number of false positives. This study introduces the new hybrid architecture development that combines the advantages of recurrent and statistical methods, the theoretical properties formalisation of gated cells for long-term memory, and the proposal of a practical approach to the model solutions’ explainability. The developed method implementation, implemented in the specialised software product form, is shown in a forensic audit. Full article

► Show Figures

Figure 1

12 pages, 1019 KB

Open AccessArticle

Blockchain and Digital Marketing: An Innovative System for Detecting Fake Comments in Search Engine Optimization Techniques and Enhancing Trust in Digital Markets

by Mouhssine Abirou, Noureddine Abghour and Zouhair Chiba

Appl. Syst. Innov. 2025, 8(5), 155; https://doi.org/10.3390/asi8050155 - 17 Oct 2025

Viewed by 287

Abstract

A significant number of digital marketers use unethical marketing methods that violate Search Engine Optimization guidelines, with the objective of deceiving engines into displaying a specific website as the top result. The practice of fake comments constitutes a violation of Search Engine Optimization [...] Read more.

A significant number of digital marketers use unethical marketing methods that violate Search Engine Optimization guidelines, with the objective of deceiving engines into displaying a specific website as the top result. The practice of fake comments constitutes a violation of Search Engine Optimization policies and is directly impeding market transparency. In addition, the absence of established standards between search engines, evaluation platforms and other trusted agencies makes exploitation easy. Therefore, in order to ensure fair competition among digital businesses, we propose a decentralized system for detecting fake comments, leveraging Blockchain technology for verification. The implementation of smart contracts as self-executing agreements will be achieved by utilizing the Ethereum network and the Truffle Suite. The Ethereum smart contracts will immutably record every comment as a transaction, eliminating any central authority. When a comment is flagged as suspicious, a digital business can trigger a verification request. Stakeholders or reviewers then vote on authenticity. Smart contracts collect these votes and issue a definitive verdict on whether the comment is fake. Full article

(This article belongs to the Special Issue Applied System Innovations Using Graph-Based Artificial Intelligence Techniques)

► Show Figures

Figure 1

60 pages, 1807 KB

Open AccessReview

Recent Advances of Artificial Intelligence Methods in PMSM Condition Monitoring and Fault Diagnosis in Elevator Systems

by Vasileios I. Vlachou, Theoklitos S. Karakatsanis and Dimitrios E. Efstathiou

Appl. Syst. Innov. 2025, 8(5), 154; https://doi.org/10.3390/asi8050154 - 16 Oct 2025

Viewed by 493

Abstract

Permanent magnet synchronous motors are the dominant technology in industrial applications such as elevator systems. Their unique advantages over induction motors give them higher energy efficiency and significant reduction in energy consumption. Accordingly, the elevator is one of the basic means of comfortable [...] Read more.

Permanent magnet synchronous motors are the dominant technology in industrial applications such as elevator systems. Their unique advantages over induction motors give them higher energy efficiency and significant reduction in energy consumption. Accordingly, the elevator is one of the basic means of comfortable and safe transportation. More generally, in elevator systems, electric motors are characterized by continuous use, increasing the risk of possible failure that may affect the operation of the system and the safety of passengers. The application of appropriate monitoring and artificial intelligence techniques contributes to the predictive maintenance of the motor and drive system. The main objective of this paper is a literature review on the application of modern monitoring methodologies using smart sensors and machine learning algorithms for early fault diagnosis and predictive maintenance generally. Thus, by exploiting the advantages and disadvantages of each method, a technique based on a multi-fault set is developed that can be integrated into an elevator control system offering desired results of immediate predictive maintenance. Full article

(This article belongs to the Section Industrial and Manufacturing Engineering)

► Show Figures

Figure 1

33 pages, 6967 KB

Open AccessArticle

LCxNet: An Explainable CNN Framework for Lung Cancer Detection in CT Images Using Multi-Optimizer and Visual Interpretability

by Noor S. Jozi and Ghaida A. Al-Suhail

Appl. Syst. Innov. 2025, 8(5), 153; https://doi.org/10.3390/asi8050153 - 15 Oct 2025

Viewed by 503

Abstract

Lung cancer, the leading cause of cancer-related mortality worldwide, necessitates better methods for earlier and more accurate detection. To this end, this study introduces LCxNet, a novel, custom-designed convolutional neural network (CNN) framework for computer-aided diagnosis (CAD) of lung cancer. The IQ-OTH/NCCD lung [...] Read more.

Lung cancer, the leading cause of cancer-related mortality worldwide, necessitates better methods for earlier and more accurate detection. To this end, this study introduces LCxNet, a novel, custom-designed convolutional neural network (CNN) framework for computer-aided diagnosis (CAD) of lung cancer. The IQ-OTH/NCCD lung CT dataset, which includes three different classes—benign, malignant, and normal—is used to train and assess the model. The framework is implemented using five optimizers, SGD, RMSProp, Adam, AdamW, and NAdam, to compare the learning behavior and performance stability. To bridge the gap between model complexity and clinical utility, we integrated Explainable AI (XAI) methods, specifically Grad-CAM for decision visualization and t-SNE for feature space analysis. With accuracy, specificity, and AUC values of 99.39%, 99.45%, and 100%, respectively, the results demonstrate that the LCxNet model outperformed the state-of-the-art models in terms of diagnostic performance. In conclusion, this study emphasizes how crucial XAI is to creating trustworthy and efficient clinical tools for the early detection of lung cancer. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

23 pages, 5276 KB

Open AccessArticle

Design and Simulation of Rotating Spray Nozzles for Greenhouse Hanging Track Spray Robots

by Siyi He, Jialin Yu and Yong Chen

Appl. Syst. Innov. 2025, 8(5), 152; https://doi.org/10.3390/asi8050152 - 14 Oct 2025

Viewed by 230

Abstract

This paper addresses deficiencies in existing spray carts and suspended sprayers regarding operational scenarios, spray coverage, versatility, and wall film thickness adjustment by designing a rail-mounted rotating nozzle application robot. Static analysis of the robot frame verifies compliance with strength and stiffness requirements. [...] Read more.

This paper addresses deficiencies in existing spray carts and suspended sprayers regarding operational scenarios, spray coverage, versatility, and wall film thickness adjustment by designing a rail-mounted rotating nozzle application robot. Static analysis of the robot frame verifies compliance with strength and stiffness requirements. Motor torque calculations ensure stable and reliable nozzle rotation. Geometric modeling derives optimal link parameters for automated nozzle angle control. ANSYS Fluent simulations characterize static spray coverage, analyzing quantitative relationships between nozzle height, angle, and spray distance. SolidWorks Motion establishes a coupled model of nozzle rotation and cart translation to obtain spray trajectories under varying speeds. Coupled Fluent simulations further evaluate wall film thickness distribution patterns under dynamic spraying conditions. The findings provide a theoretical foundation and technical reference for structural optimization and precise spraying control in greenhouse spraying robot systems. Full article

► Show Figures

Figure 1

23 pages, 4581 KB

Open AccessArticle

A Dual-Robot Digital Radiographic Inspection System for Rocket Tank Welds

by Guangbao Li, Changxing Shao, Zhiqi Wang, Yong Lu, Kenan Deng and Dong Gao

Appl. Syst. Innov. 2025, 8(5), 151; https://doi.org/10.3390/asi8050151 - 14 Oct 2025

Viewed by 321

Abstract

At present, traditional X-ray inspection is used to inspect the welds of the bottom, barrel section and short shell parts of the launch vehicle, which has the disadvantages of low automation, complicated process and low efficiency, and cannot meet the fast-paced development needs [...] Read more.

At present, traditional X-ray inspection is used to inspect the welds of the bottom, barrel section and short shell parts of the launch vehicle, which has the disadvantages of low automation, complicated process and low efficiency, and cannot meet the fast-paced development needs of multiple models at present. Moreover, the degree of digitization is low, the test results are recorded in the form of negatives, data statistics, storage and access are difficult, and the circulation efficiency is low, which is not conducive to product quality control and traceability; At the same time, it cannot adapt to and meet the needs of digital and intelligent transformation and development. In this paper, a dual-robot collaborative digital radiographic inspection system for rocket tank welds is developed by combining dual-robot control technology and digital radiographic inspection technology. The system can be directly applied to digital radiographic inspection of tank bottom, barrel section and short shell welds of multiple types of launch vehicles; meanwhile, the dual-robot path planning technology based on the dual-mode is studied. Finally, the imaging software platform based on VS and Twincat3.0 VS2015 software combined with QT upper computer is designed. Experiments show that compared with the existing traditional ray detection methods, the detection efficiency of the system is improved by 5 times, the image sensitivity reaches W14, the resolution reaches D10, and the standardized signal-to-noise ratio reaches 128, which far exceeds the requirements of process technology A, and meets the current non-destructive detection work of multi-model rocket tank welds. Full article

► Show Figures

Figure 1

20 pages, 683 KB

Open AccessArticle

Analyzing Critical Factors for the Automotive Industry’s Transition to a Circular Economy: A Multi-Attribute Decision-Making Analysis Approach

by Roxana-Mariana Nechita, Dana-Corina Deselnicu, Simona-Elena Istriţeanu and Valentina-Daniela Băjenaru

Appl. Syst. Innov. 2025, 8(5), 150; https://doi.org/10.3390/asi8050150 - 13 Oct 2025

Viewed by 384

Abstract

The automotive industry is at a critical juncture, facing increasing pressure from stringent environmental regulations, resource scarcity, and global supply chain disruptions. This study aims to identify, model, and prioritize the key factors influencing the adoption of circular economy principles in the automotive [...] Read more.

The automotive industry is at a critical juncture, facing increasing pressure from stringent environmental regulations, resource scarcity, and global supply chain disruptions. This study aims to identify, model, and prioritize the key factors influencing the adoption of circular economy principles in the automotive sector. The Decision-Making Trial and Evaluation Laboratory method was applied to analyze the interdependencies among ten critical factors. Based on a contribution from a panel of six experts from a European automotive distributor with its main operating points in France and Romania, the study reveals a clear cause-and-effect relationship among the factors. Causal factors such as organizational culture, internal integration, regulatory policies, circular business models, and R&D capacity were identified as key drivers. Conversely, factors like top management commitment, stakeholder engagement, technological capability, sustainable materials management, and ecodesign were classified as effect factors, meaning they are influenced by other variables. The key practical contribution of this research is a strategic prioritization framework for decision-makers, offering guidance on how to effectively boost key factors, either directly or indirectly, to achieve a more efficient alignment with circular economy principles. By strategically leveraging these interdependencies, organizations can trigger a positive chain reaction, leading to improved performance in the dependent effect factors and ultimately accelerating the transition to a circular economy. Full article

► Show Figures

Figure 1

28 pages, 7176 KB

Open AccessArticle

Resilience Oriented Distribution System Service Restoration Considering Overhead Power Lines Affected by Hurricanes

by Kehkashan Fatima, Hussain Shareef and Flavio Bezerra Costa

Appl. Syst. Innov. 2025, 8(5), 149; https://doi.org/10.3390/asi8050149 - 9 Oct 2025

Viewed by 438

Abstract

In recent years, there has been an increase in the frequency of severe weather events (like hurricanes). These events are responsible for most power outages in power distribution systems (PDSs). Particularly susceptible to storms are overhead PDSs. In this study, the dynamic Bayesian [...] Read more.

In recent years, there has been an increase in the frequency of severe weather events (like hurricanes). These events are responsible for most power outages in power distribution systems (PDSs). Particularly susceptible to storms are overhead PDSs. In this study, the dynamic Bayesian network (DBN)-based failure model was developed for different hurricane scenarios to predict the line failure of overhead lines. Based on the outcomes of the DBN model, a service restoration model was formulated to maximize restored loads and minimize power losses using Particle Swarm Optimization (PSO)-based distributed generation (DG) integration and system reconfiguration. Three different case studies based on the IEEE 33 bus system were conducted. The overhead line failure prediction and service restoration model findings were further used to calculate resilience metrics. With reconfiguration the load restored from 90.3% to 100% for Case 1 and from 34.994% to 80.35% for Case 2. However, for Case 3, reconfiguration alone was not sufficient to show any improvement in performance. On the other hand, DG integration successfully restored load to 100% in all three cases. These results demonstrated that the combined DBN-based failure modeling and PSO-driven optimal restoration strategy under hurricane-induced disruptions can effectively strengthen system resilience. Full article

► Show Figures

Figure 1

16 pages, 661 KB

Open AccessArticle

A Two-Layer Model for Complex Multi-Criteria Decision-Making and Its Application in Institutional Research

by Yinghui Zhou and Atsushi Asano

Appl. Syst. Innov. 2025, 8(5), 148; https://doi.org/10.3390/asi8050148 - 7 Oct 2025

Viewed by 474

Abstract

Complex decision-making often involves numerous alternatives and diverse criteria, making it difficult to set clear priorities under resource constraints. This study proposes a two-layer hierarchical decision model that structures the process into sequential stages: the first layer narrows the alternatives according to strategic [...] Read more.

Complex decision-making often involves numerous alternatives and diverse criteria, making it difficult to set clear priorities under resource constraints. This study proposes a two-layer hierarchical decision model that structures the process into sequential stages: the first layer narrows the alternatives according to strategic considerations, and the second layer re-evaluates the shortlisted options based on feasibility. This layered design clarifies the decision path and enhances interpretability compared to single-layer approaches. To demonstrate its practical value, the model is applied to an institutional research case in higher education, implemented with the entropy weight method (EWM) for weighting and TOPSIS for ranking. The results demonstrate that it supports transparent and resource-aware planning for performance improvement, while being scalable to multi-layer structure to accommodate diverse organizational needs and varying levels of complexity. Full article

► Show Figures

Figure 1

32 pages, 12079 KB

Open AccessArticle

Fault Diagnosis in Internal Combustion Engines Using Artificial Intelligence Predictive Models

by Norah Nadia Sánchez Torres, Joylan Nunes Maciel, Thyago Leite de Vasconcelos Lima, Mario Gazziro, Abel Cavalcante Lima Filho, João Paulo Pereira do Carmo and Oswaldo Hideo Ando Junior

Appl. Syst. Innov. 2025, 8(5), 147; https://doi.org/10.3390/asi8050147 - 30 Sep 2025

Viewed by 829

Abstract

The growth of greenhouse gas emissions, driven by the use of internal combustion engines (ICE), highlights the urgent need for sustainable solutions, particularly in the shipping sector. Non-invasive predictive maintenance using acoustic signal analysis has emerged as a promising strategy for fault diagnosis [...] Read more.

The growth of greenhouse gas emissions, driven by the use of internal combustion engines (ICE), highlights the urgent need for sustainable solutions, particularly in the shipping sector. Non-invasive predictive maintenance using acoustic signal analysis has emerged as a promising strategy for fault diagnosis in ICEs. In this context, the present study proposes a hybrid Deep Learning (DL) model and provides a novel publicly available dataset containing real operational sound samples of ICEs, labeled across 12 distinct fault subclasses. The methodology encompassed dataset construction, signal preprocessing using log-mel spectrograms, and the evaluation of several Machine Learning (ML) and DL models. Among the evaluated architectures, the proposed hybrid model, BiGRUT (Bidirectional GRU + Transformer), achieved the best performance, with an accuracy of 97.3%. This architecture leverages the multi-attention capability of Transformers and the sequential memory strength of GRUs, enhancing robustness in complex fault scenarios such as combined and mechanical anomalies. The results demonstrate the superiority of DL models over traditional ML approaches in acoustic-based ICE fault detection. Furthermore, the dataset and hybrid model introduced in this study contribute toward the development of scalable real-time diagnostic systems for sustainable and intelligent maintenance in transportation systems. Full article

► Show Figures

Figure 1

35 pages, 12402 KB

Open AccessArticle

A Multi-Teacher Knowledge Distillation Framework with Aggregation Techniques for Lightweight Deep Models

by Ahmed Hamdi, Hassan N. Noura and Joseph Azar

Appl. Syst. Innov. 2025, 8(5), 146; https://doi.org/10.3390/asi8050146 - 30 Sep 2025

Viewed by 502

Abstract

Knowledge Distillation (KD) is a machine learning technique in which a compact student model learns to replicate the performance of a larger teacher model by mimicking its output predictions. Multi-Teacher Knowledge Distillation extends this paradigm by aggregating knowledge from multiple teacher models to [...] Read more.

Knowledge Distillation (KD) is a machine learning technique in which a compact student model learns to replicate the performance of a larger teacher model by mimicking its output predictions. Multi-Teacher Knowledge Distillation extends this paradigm by aggregating knowledge from multiple teacher models to improve generalization and robustness. However, effectively integrating outputs from diverse teachers, especially in the presence of noise or conflicting predictions, remains a key challenge. In this work, we propose a Multi-Round Parallel Multi-Teacher Distillation (MPMTD) that systematically explores and combines multiple aggregation techniques. Specifically, we investigate aggregation at different levels, including loss-based and probability-distribution-based fusion. Our framework applies different strategies across distillation rounds, enabling adaptive and synergistic knowledge transfer. Through extensive experimentation, we analyze the strengths and weaknesses of individual aggregation methods and demonstrate that strategic sequencing across rounds significantly outperforms static approaches. Notably, we introduce the Byzantine-Resilient Probability Distribution aggregation method applied for the first time in a KD context, which achieves state-of-the-art performance, with an accuracy of 99.29% and an F1-score of 99.27%. We further identify optimal configurations in terms of the number of distillation rounds and the ordering of aggregation strategies, balancing accuracy with computational efficiency. Our contributions include (i) the introduction of advanced aggregation strategies into the KD setting, (ii) a systematic evaluation of their performance, and (iii) practical recommendations for real-world deployment. These findings have significant implications for distributed learning, edge computing, and IoT environments, where efficient and resilient model compression is essential. Full article

► Show Figures

Figure 1

35 pages, 5864 KB

Open AccessArticle

Risk-Constrained Multi-Objective Deep Reinforcement Learning for AGV Path Planning in Rail Transit

by Zihan Yang and Huiyu Xiang

Appl. Syst. Innov. 2025, 8(5), 145; https://doi.org/10.3390/asi8050145 - 30 Sep 2025

Viewed by 460

Abstract

Sensor-rich Automated Guided Vehicles (AGVs) are increasingly deployed in logistics, yet large fleets relying on fixed tracks face high maintenance costs and frequent route conflicts. This study targets rail-based material handling and proposes an end-to-end multi-AGV navigation pipeline under realistic operational constraints. A [...] Read more.

Sensor-rich Automated Guided Vehicles (AGVs) are increasingly deployed in logistics, yet large fleets relying on fixed tracks face high maintenance costs and frequent route conflicts. This study targets rail-based material handling and proposes an end-to-end multi-AGV navigation pipeline under realistic operational constraints. A conflict-aware global planner, extended from the A* algorithm, generates feasible routes, while a multi-sensor perception stack integrates LiDAR and camera data to distinguish moving AGVs, static obstacles, and task targets. Based on this perception, a Deep Q-Network (DQN) policy with a tailored reward function enables real-time dynamic obstacle avoidance in complex traffic. Simulation results demonstrate that, compared with the Artificial Potential Field (APF) baseline, the proposed GG-DRL approach reduces collisions by ~70%, lowers planning time by 25–30%, shortens paths by 10–15%, and improves smoothness by 20–25%. On the Maze Benchmark Map, GG-DRL surpasses classical planners (e.g., RRT) and deep RL baselines (e.g., DDPG) in path quality, computation, and avoidance behavior, achieving an average path length of 81.12, computation time of 11.94 s, 5.2 avoidance maneuvers, and smoothness of 0.86. Robustness is maintained as a dynamic obstacles scale up to 30. These findings confirm that combining multi-sensor fusion with deep reinforcement learning enhances AGV safety, efficiency, and reliability, with broad potential for intelligent railway logistics. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

37 pages, 905 KB

Open AccessReview

Application of Fuzzy Logic Techniques in Solar Energy Systems: A Review

by Siviwe Maqekeni, KeChrist Obileke, Odilo Ndiweni and Patrick Mukumba

Appl. Syst. Innov. 2025, 8(5), 144; https://doi.org/10.3390/asi8050144 - 30 Sep 2025

Viewed by 542

Abstract

Fuzzy logic has been applied to a wide range of problems, including process control, object recognition, image and signal processing, prediction, classification, decision-making, optimization, and time series analysis. These apply to solar energy systems. Though experts in renewable energy prefer fuzzy logic techniques, [...] Read more.

Fuzzy logic has been applied to a wide range of problems, including process control, object recognition, image and signal processing, prediction, classification, decision-making, optimization, and time series analysis. These apply to solar energy systems. Though experts in renewable energy prefer fuzzy logic techniques, their contribution to the decision-making process of solar energy systems lies in the possibility of illustrating risk factors and introducing the concepts of linguistic variables of data from solar energy applications. In solar energy systems, the primary beneficiaries and audience of the fuzzy logic techniques are solar energy policy makers, as it concerns decision-making models, ranking of criteria or weights, and assessment of the potential location of the installation of solar energy plants, depending on the case. In a real-world scenario, fuzzy logic allows easy and efficient controller configuration in a non-linear control system, such as a solar panel. This study attempts to review the role and contribution of fuzzy logic in solar energy based on its applications. The findings from the review revealed that the fuzzy logic application identifies and detects faults in solar energy systems as well as in the optimization of energy output and the location of solar energy plants. In addition, fuzzy model (predicting), hybrid model (simulating performance), and multi-criteria decision-making (MCDM) are components of fuzzy logic techniques. As the review indicated, these are useful as a solution to the challenges of solar energy systems. Importantly, the integration and incorporation of fuzzy logic and neural networks should be recommended for the efficient and effective performance of solar energy systems. Full article

► Show Figures

Figure 1

21 pages, 2749 KB

Open AccessArticle

Performance Analysis of an Optical System for FSO Communications Utilizing Combined Stochastic Gradient Descent Optimization Algorithm

by Ilya Galaktionov and Vladimir Toporovsky

Appl. Syst. Innov. 2025, 8(5), 143; https://doi.org/10.3390/asi8050143 - 30 Sep 2025

Viewed by 421

Abstract

Wavefront aberrations caused by thermal flows or arising from the quality of optical components can significantly impair wireless communication links. Such aberrations may result in an increased error rate in the received signal, leading to data loss in laser communication applications. In this [...] Read more.

Wavefront aberrations caused by thermal flows or arising from the quality of optical components can significantly impair wireless communication links. Such aberrations may result in an increased error rate in the received signal, leading to data loss in laser communication applications. In this study, we explored a newly developed combined stochastic gradient descent optimization algorithm aimed at compensating for optical distortions. The algorithm we developed exhibits linear time and space complexity and demonstrates low sensitivity to variations in input parameters. Furthermore, its implementation is relatively straightforward and does not necessitate an in-depth understanding of the underlying system, in contrast to the Stochastic Parallel Gradient Descent (SPGD) method. In addition, a developed switch-mode approach allows us to use a stochastic component of the algorithm as a rapid, rough-tuning mechanism, while the gradient descent component is used as a slower, more precise fine-tuning method. This dual-mode operation proves particularly advantageous in scenarios where there are no rapid dynamic wavefront distortions. The results demonstrated that the proposed algorithm significantly enhanced the total collected power of the beam passing through the 10 μm diaphragm that simulated a 10 μm fiber core, increasing it from 0.33 mW to 2.3 mW. Furthermore, the residual root mean square (RMS) aberration was reduced from 0.63 μm to 0.12 μm, which suggests a potential improvement in coupling efficiency from 0.1 to 0.6. Full article

(This article belongs to the Section Information Systems)

► Show Figures

Figure 1

28 pages, 1463 KB

Open AccessArticle

Strategic Management Knowledge Map via BERTopic (1980–2025): Evolution, Integration, and Application

by Kuei-Kuei Lai, Chih-Wen Hsiao and Yu-Jin Hsu

Appl. Syst. Innov. 2025, 8(5), 142; https://doi.org/10.3390/asi8050142 - 29 Sep 2025

Viewed by 599

Abstract

Problem: Amid digital disruption and the cross-fertilization of RBV, DCV, and KBV, strategic management knowledge has grown fragmented with blurred boundaries. Conventional mapping (citation/co-word, LDA) lacks semantic and temporal resolution, obscuring overlaps, divergences, and turning points and hindering links to actionable indicators (e.g., [...] Read more.

Problem: Amid digital disruption and the cross-fertilization of RBV, DCV, and KBV, strategic management knowledge has grown fragmented with blurred boundaries. Conventional mapping (citation/co-word, LDA) lacks semantic and temporal resolution, obscuring overlaps, divergences, and turning points and hindering links to actionable indicators (e.g., the Balanced Scorecard). Hence, an integrated, semantically faithful, time-stamped map is needed to bridge research and operational metrics. Gap: Prior maps rely on citation/co-word signals, miss textual meaning, and treat RBV/DCV/KBV in isolation—lacking a theory-aligned, time-stamped, manager-oriented synthesis. Objectives: This study aims to (1) reveal how RBV, DCV, and KBV evolve and interrelate over time; (2) produce an integrated, semantically grounded map; and (3) translate selected themes into actionable managerial indicators. Method: We analyzed 25,907 WoS articles (1980–2025) with BERTopic (Sentence-BERT + UMAP + HDBSCAN + c-TF-IDF). We used an RBV/DCV/KBV lexicon to guide retrieval/interpretation (not to constrain modeling). We discovered 230 topics, retained 33 via coherence (C_V), and benchmarked them against LDA. Key findings: A concise set of 33 high-quality themes with a higher C_V than LDA on this corpus was established. A Fish-Scale view (overlapping subfields across economics, management, sociology) that clarifies RBV–DCV–KBV intersections was achieved. Era-sliced prevalence shows how themes emerge and recombine over 1980–2025. Selected themes mapped to Balanced Scorecard (BSC) indicators linking capabilities → processes → customer outcomes → financial results. Contribution: A clear, time-aware synthesis of RBV–DCV–KBV and a scalable, reproducible pipeline for structuring fragmented theory landscapes are presented in this study—bridging scholarly integration with managerial application via BSC mapping. Full article

(This article belongs to the Topic Social Sciences and Intelligence Management, 2nd Volume)

► Show Figures

Figure 1

19 pages, 2248 KB

Open AccessArticle

A Platform for Machine Learning Operations for Network Constrained Far-Edge Devices

by Calum McCormack and Imene Mitiche

Appl. Syst. Innov. 2025, 8(5), 141; https://doi.org/10.3390/asi8050141 - 28 Sep 2025

Viewed by 515

Abstract

Machine Learning (ML) models developed for the Edge have seen a massive uptake in recent years, with many types of predictive analytics, condition monitoring and pre-emptive fault detection developed and in-use on Internet of Things (IoT) systems serving industrial power generators, environmental monitoring [...] Read more.

Machine Learning (ML) models developed for the Edge have seen a massive uptake in recent years, with many types of predictive analytics, condition monitoring and pre-emptive fault detection developed and in-use on Internet of Things (IoT) systems serving industrial power generators, environmental monitoring systems and more. At scale, these systems can be difficult to manage and keep upgraded, especially those devices that are deployed in far-Edge networks with unreliable networking. This paper presents a simple and novel platform architecture for deployment and management of ML at the Edge for increasing model and device reliability by reducing downtime and access to new model versions via the ability to manage models from both Cloud and Edge. This platform provides an Edge ML Operations “Mirror” that replicates and minimises cloud MLOps systems to provide reliable delivery and retraining of models at the network Edge, solving many problems associated with both Cloud-first and Edge networks. The paper explores and explains the architecture and components of the system, offering a prototype system that was evaluated by measuring time to deploy models with regard to differing network instabilities in a simulated environment to highlight the necessity for local management and federated training of models as a secondary function to Cloud model management. This architecture could be utilised by researchers to improve the deployment, recording and management of ML experiments on the Edge. Full article

(This article belongs to the Topic Federated Edge Intelligence for Next Generation AI Systems)

► Show Figures

Figure 1

15 pages, 3480 KB

Open AccessArticle

Graphics-Guided Interactive Farmland Layout Design

by Guanlin Liu and Huijun Yang

Appl. Syst. Innov. 2025, 8(5), 140; https://doi.org/10.3390/asi8050140 - 25 Sep 2025

Viewed by 536

Abstract

The spatial layout of farmland involves coordinated planning across diverse functional zones. Irregular land boundaries and functional demands pose challenges to traditional CAD-based workflows and general optimization algorithms. To address these limitations, we propose an interactive farmland layout system based on the Graphic-Guided [...] Read more.

The spatial layout of farmland involves coordinated planning across diverse functional zones. Irregular land boundaries and functional demands pose challenges to traditional CAD-based workflows and general optimization algorithms. To address these limitations, we propose an interactive farmland layout system based on the Graphic-Guided Evolutionary Layout (GGEL) algorithm. GGEL not only introduces a graph-based spatial pruning and encoding strategy to improve search efficiency, but also performs real-time spatial overlap detection based on functional region boundaries to ensure layout feasibility. Additionally, an interactive module enables real-time user customization to accommodate specific planning needs. Experimental results demonstrate that the system can efficiently generate complete multi-region layouts, significantly enhancing design productivity. A user study with 20 agricultural park experts confirms the system’s usability and effectiveness. This study highlights the potential of integrating evolutionary algorithms with topological graph representations to address the complex spatial planning requirements of digital agricultural parks. Full article

(This article belongs to the Section Human-Computer Interaction)

► Show Figures

Figure 1

19 pages, 867 KB

Open AccessArticle

Development of a Solution for Smart Home Management System Selection Based on User Needs

by Daiva Stanelytė, Birutė Rataitė, Algimantas Andriušis, Aleksas Narščius, Gintaras Kučinskas and Jelena Dikun

Appl. Syst. Innov. 2025, 8(5), 139; https://doi.org/10.3390/asi8050139 - 24 Sep 2025

Viewed by 436

Abstract

The complexity of smart home technologies and the need for personalized energy efficiency solutions highlight the importance of user-oriented decision-support tools. This study presents a Smart Home Management System (SHMS) selection solution that combines a web-based dashboard, a mobile application, and a relational [...] Read more.

The complexity of smart home technologies and the need for personalized energy efficiency solutions highlight the importance of user-oriented decision-support tools. This study presents a Smart Home Management System (SHMS) selection solution that combines a web-based dashboard, a mobile application, and a relational database. A 54-question structured questionnaire was designed to capture user requirements, and four alternatives—KNX, JUNG Home, LB Management, and eNet Smart Home—were compared using the Simple Additive Weighting (SAW) method. Evaluation criteria included installation complexity, communication technology, integration and control capabilities, and user experience. The system was implemented with Next.js, React Native, and Post-greSQL, ensuring flexibility, scalability, and secure data management. Preliminary evaluation with specialists (system integrators, architects, designers) and students confirmed the coherence of the questionnaire, the adequacy of criteria, and the clarity of recommendations. Results showed that the tool improves user engagement, reduces decision-making uncertainty, and supports the adoption of energy-efficient residential solutions. The study’s main limitation is the small test sample, which will be expanded in future large-scale validation. Planned improvements include interactive product comparisons, cost estimation, adaptive questionnaire logic, and 3D visualizations. Overall, the system bridges the gap between technical SHMS solutions and user-oriented decision-making, offering practical and academic value. Full article

(This article belongs to the Special Issue Feature Papers in the ‘Applied Systems on Educational Innovations and Emerging Technologies’ Section)

13 pages, 3006 KB

Open AccessArticle

A Novel Controller for Fuel Cell Generators Based on CAN Bus

by Ching-Hsu Chan, Fuh-Liang Wen, Chu-Po Wen and Kevin Karindra Putra Pradana

Appl. Syst. Innov. 2025, 8(5), 138; https://doi.org/10.3390/asi8050138 - 24 Sep 2025

Viewed by 509

Abstract

The novel design and modular implementation of a distributed control system for a fuel cell generator, aimed at monitoring and actuation, are presented. Two ESP32 NodeMCU microcontrollers and MCP2515 modules are used for the controller area network (CAN) bus communication protocol. To compare [...] Read more.

The novel design and modular implementation of a distributed control system for a fuel cell generator, aimed at monitoring and actuation, are presented. Two ESP32 NodeMCU microcontrollers and MCP2515 modules are used for the controller area network (CAN) bus communication protocol. To compare this setup with a traditional battery management system (BMS), small rated-power fuel cell generators were connected individually via the CAN bus to form a larger stacked output. An RFID interface was introduced into the CAN bus system to enhance its applicability in stacked fuel cells, without interfering with original message frames, arbitration mechanisms, or CRC efficiency across various sectors. Additionally, to provide a clearer understanding of the system’s features and functions, a PC-based logic analyzer was employed as an analytical tool to monitor and analyze data transmitted over the CAN bus. Comprehensive insights into the system’s performance are supported by logic analysis of its complex applications in series-connected fuel cells. The advantages of the RFID-based CAN bus are further enhanced by modern communication protocols, offering greater scalability and flexibility, with potential applications in industrial automation, autonomous vehicles, and smart green power grids. Full article

(This article belongs to the Topic Application of IoT on Manufacturing, Communication and Engineering)

► Show Figures

Figure 1

21 pages, 2107 KB

Open AccessReview

Digitalisation in the Context of Industry 4.0 and Industry 5.0: A Bibliometric Literature Review and Visualisation

by Zsolt Buri and Judit T. Kiss

Appl. Syst. Innov. 2025, 8(5), 137; https://doi.org/10.3390/asi8050137 - 23 Sep 2025

Viewed by 649

Abstract

This study examines industrial digitalization, with a particular focus on the transformation from Industry 4.0 to Industry 5.0. The research is based on a database of 1441 Scopus-indexed articles, which forms the basis of a systematic literature review and bibliometric network analysis. The [...] Read more.

This study examines industrial digitalization, with a particular focus on the transformation from Industry 4.0 to Industry 5.0. The research is based on a database of 1441 Scopus-indexed articles, which forms the basis of a systematic literature review and bibliometric network analysis. The articles were ranked using Global Citation Score (GCS), followed by Co-Coupling Network (CCN) within VosViewer, the method to create arrays. The arrays were analyzed based on the connection strengths of the citations in them. Next, we performed Burst Detection using the CiteSpace app. Finally, the most relevant keywords, determined in the Burst Detection, were used for Co-Occurrence Network (CONK), with which we could create new arrays and analyze them. By connecting the various, fragmented scientific findings, our results highlight that digital twins, artificial intelligence, supply chain resilience and the Internet of Things are the focus of Industry 4.0, i.e., the technological side is dominant. In contrast, Industry 5.0 places employees at the center. It also emphasizes the analysis of human–machine interaction and the importance of green digital sustainability. The results provide a comprehensive picture of how decision-makers, researchers, and professionals can interpret a changing mindset and apply it as practical advice. Full article

► Show Figures

Figure 1

16 pages, 2816 KB

Open AccessArticle

Hardware-Encrypted System for Storage of Collected Data Based on Reconfigurable Architecture

by Vasil Gatev, Valentin Mollov and Adelina Aleksieva-Petrova

Appl. Syst. Innov. 2025, 8(5), 136; https://doi.org/10.3390/asi8050136 - 22 Sep 2025

Viewed by 421

Abstract

This submission is focused on the implementation of a system that acquires data from various types of sensors and securely stores them after encryption on a chip with a reconfigurable architecture. The system has the unique capability of encrypting the input data with [...] Read more.

This submission is focused on the implementation of a system that acquires data from various types of sensors and securely stores them after encryption on a chip with a reconfigurable architecture. The system has the unique capability of encrypting the input data with a single secret cryptographic key, which is stored only inside the hardware of the system itself, so the key remains unrecognizable upon completion of the system synthesis for any unauthorized user. Being stored as a part of the whole system architecture, the cryptographic key cannot be attained. It is not stored separately on the system RAM or any other supported memory, making the collected data fully protected. The reported work shows a data acquisition system which measures temperature with a high level of precision, transforms it to degrees Celsius, stores the collected data, and transfers them via serial interface when requested. Before storage, the data are encrypted with a 256-bit key, applying the AES algorithm. The data which are stored in the system memory and sent as UART packets towards the main computer do not include the cryptographic key in the data stream, so it is impossible for it to be retrieved from them. We show the flexibility of such kinds of data acquisition systems for sensing different types of signals, emphasizing secure storage and transferring, including data from meteorological sensors or highly confidential or biometrical data. Full article

► Show Figures

Figure 1

27 pages, 7440 KB

Open AccessArticle

Buffer with Dropping Function and Correlated Packet Lengths

by Andrzej Chydzinski and Blazej Adamczyk

Appl. Syst. Innov. 2025, 8(5), 135; https://doi.org/10.3390/asi8050135 - 19 Sep 2025

Viewed by 499

Abstract

We analyze a model of the packet buffer in which a new packet can be discarded with a probability connected to the buffer occupancy through an arbitrary dropping function. Crucially, it is assumed that packet lengths can be correlated in any way and [...] Read more.

We analyze a model of the packet buffer in which a new packet can be discarded with a probability connected to the buffer occupancy through an arbitrary dropping function. Crucially, it is assumed that packet lengths can be correlated in any way and that the interarrival time has a general distribution. From an engineering perspective, such a model constitutes a generalization of many active buffer management algorithms proposed for Internet routers. From a theoretical perspective, it generalizes a class of finite-buffer models with the tail-drop discarding policy. The contributions include formulae for the distribution of buffer occupancy and the average buffer occupancy, at arbitrary times and also in steady state. The formulae are illustrated with numerical calculations performed for various dropping functions. The formulae are also validated via discrete-event simulations. Full article

(This article belongs to the Section Applied Mathematics)

► Show Figures

Figure 1

18 pages, 952 KB

Open AccessArticle

Advanced Vehicle Electrical System Modelling for Software Solutions on Manufacturing Plants: Proposal and Applications

by Adrià Bosch Serra, Juan Francisco Blanes Noguera, Luis Ruiz Matallana, Carlos Álvarez Baldo and Joan Porcar Rodado

Appl. Syst. Innov. 2025, 8(5), 134; https://doi.org/10.3390/asi8050134 - 17 Sep 2025

Viewed by 559

Abstract

Mass customisation in the automotive industry has exploded the number of wiring harness variants that must be assembled, tested and repaired on the shop floor. Existing CAD or schematic formats are too heavy and too coarse-grained to drive in-line, per-VIN validation, while supplier [...] Read more.

Mass customisation in the automotive industry has exploded the number of wiring harness variants that must be assembled, tested and repaired on the shop floor. Existing CAD or schematic formats are too heavy and too coarse-grained to drive in-line, per-VIN validation, while supplier documentation is heterogeneous and often incomplete. This paper presents a pin-centric, two-tier graph model that converts raw harness tables into a machine-readable, wiring-aware digital twin suitable for real-time use in manufacturing plants. All physical and logical artefacts—pins, wires, connections, paths and circuits—are represented as nodes, and a dual-store persistence strategy separates attribute-rich JSON documents from a lightweight NetworkX property graph. The architecture supports dozens of vehicle models and engineering releases without duplicating data, and a decentralised validation pipeline enforces both object-level and contextual rules, reducing initial domain violations from eight to zero and eliminating fifty-two circuit errors in three iterations. The resulting platform graph is generated in 0.7 s and delivers 100% path-finding accuracy. Deployed at Ford’s Almussafes plant, the model already underpins launch-phase workload mitigation, interactive visualisation and early design error detection. Although currently implemented in Python 3.11 and lacking quantified production KPIs, the approach establishes a vendor-agnostic data standard and lays the groundwork for self-aware manufacturing: future work will embed real-time validators on the line, stream defect events back into the graph and couple the wiring layer with IoT frameworks for autonomous repair and optimisation. Full article

(This article belongs to the Section Information Systems)

► Show Figures

Figure 1

15 pages, 2399 KB

Open AccessArticle

Development of a Mobile Health Monitoring and Alert Application for Agricultural Workers

by Omer Oztoprak and Ji-Chul Ryu

Appl. Syst. Innov. 2025, 8(5), 133; https://doi.org/10.3390/asi8050133 - 15 Sep 2025

Viewed by 763

Abstract

The health and safety of agricultural workers are critical concerns due to their exposure to extreme environmental conditions, physically demanding tasks, and limited access to immediate medical assistance. This study presents the design and development of a novel smartphone application that integrates multiple [...] Read more.

The health and safety of agricultural workers are critical concerns due to their exposure to extreme environmental conditions, physically demanding tasks, and limited access to immediate medical assistance. This study presents the design and development of a novel smartphone application that integrates multiple wearable physiological sensors—a fingertip pulse oximeter, a skin patch thermometer, and an inertial measurement unit (IMU)—via Bluetooth Low Energy (BLE) technology for real-time health monitoring and alert notifications. Unlike many existing platforms, the proposed system offers direct access to raw sensor data, modular multi-sensor integration, and a scalable software framework based on the Model–View–ViewModel (MVVM) architecture with Jetpack Compose for a responsive user interface. Experimental results demonstrated stable BLE connections, accurate extraction of oxygen saturation, heart rate, body temperature, and trunk inclination data, as well as reliable real-time alerts when the system detects anomalies based on predetermined thresholds. The system also incorporates automatic reconnection mechanisms to maintain continuous monitoring. Beyond agriculture, the proposed framework can be adapted to broader occupational safety domains, with future improvements focusing on additional sensors, redundant sensing, cloud-based data storage, and large-scale field validation. Full article

(This article belongs to the Section Medical Informatics and Healthcare Engineering)

► Show Figures

Figure 1

17 pages, 6119 KB

Open AccessArticle

Optimization of Elevator Standby Scheduling Strategy in Smart Buildings

by I-Ning Tsai, You-Xuan Wu, Yueh-Hsuan Huang, Yu-Chen Chen and Jian-Jiun Ding

Appl. Syst. Innov. 2025, 8(5), 132; https://doi.org/10.3390/asi8050132 - 15 Sep 2025

Viewed by 956

Abstract

Elevator Group Control Systems (EGCSs) play a key role in managing the passenger flow and consumption of energy in modern buildings. However, existing EGCS algorithms are typically only applied to real-time passenger calls, which does not take the long-term statistics of passenger requirement [...] Read more.

Elevator Group Control Systems (EGCSs) play a key role in managing the passenger flow and consumption of energy in modern buildings. However, existing EGCS algorithms are typically only applied to real-time passenger calls, which does not take the long-term statistics of passenger requirement into account. To address this gap, we propose a standby strategy that proactively repositioning idle elevators even if there is no passenger call. It calculates a combined score that balances the expected waiting time and the energy consumption to determine the optimal standby floors for idle elevators. We implement this strategy on a simple baseline dispatcher using the closest car algorithm and introduce tunable parameters to adjust the standby behavior. Experiments on mid-rise and high-rise building scenarios show that the standby strategy significantly reduces the average waiting time for passengers by more than 24% in both cases. Moreover, because this strategy operates independently of the core dispatcher, it can be combined with existing EGCS algorithms to further improve waiting time without compromising core energy optimizations. These findings demonstrate that proactive standby repositioning is an effective complementary approach for next-generation elevator control systems and offers a practical way to reduce waiting times under realistic office building traffic conditions. Full article

(This article belongs to the Topic Application of IoT on Manufacturing, Communication and Engineering)

► Show Figures

Figure 1

11 pages, 711 KB

Open AccessArticle

Integrating Machine Learning and Sustainability in Nonwoven Production: A Case Study Using the FOREST Framework

by Rosario Othen, Steven Macpherson, Christian Möbitz and Thomas Gries

Appl. Syst. Innov. 2025, 8(5), 131; https://doi.org/10.3390/asi8050131 - 12 Sep 2025

Viewed by 591

Abstract

The environmental impact of industrial processes, especially regarding CO₂ emissions, requires innovative tools to monitor and optimize resource consumption. This study presents a data-driven approach for a carded nonwoven production line, aiming to support integration into the sustainability framework FOREST (Framework for [...] Read more.

The environmental impact of industrial processes, especially regarding CO₂ emissions, requires innovative tools to monitor and optimize resource consumption. This study presents a data-driven approach for a carded nonwoven production line, aiming to support integration into the sustainability framework FOREST (Framework for Resource, Energy, Sustainability Treatment). Real process data from a pilot line were pre-processed, analysed, and used to train machine learning models to predict energy consumption across multiple production stages. Using techniques such as recursive feature elimination and SHAP value interpretation, the most influential parameters for each process step were identified. Extra Trees Regression proved to be the most accurate and explainable model across all scenarios. The results allow real-time estimation of the Product Carbon Footprint (PCF) based on process parameters and provide a foundation for energy optimization in nonwoven manufacturing. Full article

► Show Figures

Figure 1

34 pages, 9816 KB

Open AccessArticle

Residential Load Flow Modeling and Simulation

by Nikola Vojnović, Vladan Krsman, Jovana Vidaković, Milan Vidaković, Željko Popović, Dragan Pejić and Đorđe Novaković

Appl. Syst. Innov. 2025, 8(5), 130; https://doi.org/10.3390/asi8050130 - 12 Sep 2025

Viewed by 627

Abstract

In recent years, home energy management systems (HEMSs) have emerged as critical components within the concept of smart cities and grids. Within HEMSs, load flow analysis represents one of the fundamental tools for smart grid studies, forming the basis for a wide range [...] Read more.

In recent years, home energy management systems (HEMSs) have emerged as critical components within the concept of smart cities and grids. Within HEMSs, load flow analysis represents one of the fundamental tools for smart grid studies, forming the basis for a wide range of advanced applications, including state estimation, fault diagnosis, and optimal power flow computation. To achieve high levels of load flow accuracy and reliability, HEMSs must incorporate detailed models of all electrical elements found in modern residential units, including appliances, wiring, and energy resources. This paper proposes a load flow solution for smart home networks, featuring detailed models of wiring, appliances, and on-site generation systems. Moreover, a detailed appliance model derived from smart meter data, capable of representing both static-load devices and complex appliances with time-varying consumption profiles, is introduced. Additionally, a measurement-based validation of residential electrical wiring model is presented. The proposed models and calculation procedures have been verified by comparing the simulated results with the literature, yielding a deviation of approximately 0.7%. Analyses of a large-scale network have shown that this approach is up to six times faster compared to state-of-the-art procedures. The developed solution demonstrates practical applicability for use in industry-grade smart power management software. Full article

► Show Figures

Figure 1

17 pages, 5540 KB

Open AccessArticle

Enhancing Axial Flow in Hydrokinetic Turbines via Multi-Slot Diffuser Design: A Computational Study

by Daniel Sanin-Villa, Jorge Sierra-Del Rio, Diego Hincapié Zuluaga and Steven Galvis-Holguin

Appl. Syst. Innov. 2025, 8(5), 129; https://doi.org/10.3390/asi8050129 - 11 Sep 2025

Cited by 1 | Viewed by 538

Abstract

Straight-walled diffusers can boost the power density of horizontal-axis hydrokinetic turbines (HKTs), but are prone to boundary layer separation when the divergence angle is too large. We perform a systematic factorial study of three diffuser configurations, slotless, mid-length single-slot, and outlet-slot with dual [...] Read more.

Straight-walled diffusers can boost the power density of horizontal-axis hydrokinetic turbines (HKTs), but are prone to boundary layer separation when the divergence angle is too large. We perform a systematic factorial study of three diffuser configurations, slotless, mid-length single-slot, and outlet-slot with dual divergence angles, using a two-dimensional, transient SST k–

ω

Reynolds-averaged Navier–Stokes model validated against wind tunnel data (maximum error 6.4%). Eight geometries per configuration are generated through a

2^{3}

Design of Experiments with variation in the divergence angle, flange or slot position, and inlet section. The optimal outlet-slot design re-energises the boundary layer, shortens the recirculation zone by more than 50%, and raises the mean axial velocity along the diffuser centreline by 12.6% compared with an equally compact slotless diffuser, and by 42.6% relative to an open flow without a diffuser. Parametric analysis shows that the slot position in the radial (Y) direction and the first divergence angle have the strongest influence on velocity augmentation. In contrast, the flange angle and axial slot location (X) are second-order effects. The results provide fabrication-friendly guidelines, restricted to straight walls and a single slot, that are capable of improving HKT performance in shallow or remote waterways where complex curved diffusers are impractical. The study also identifies key geometric and turbulence model sensitivities that should be addressed in future three-dimensional and multi-slot investigations. Full article

► Show Figures

Figure 1

23 pages, 1584 KB

Open AccessArticle

Image-Based Formalization of Tabular Data for Threshold-Based Prediction of Hospital Stay Using Convolutional Neural Networks: An Intelligent Decision Support System Applied in COPD

by Alberto Pinheira, Manuel Casal-Guisande, Julia López-Canay, Alberto Fernández-García, Rafael Golpe, Cristina Represas-Represas, María Torres-Durán, Jorge Cerqueiro-Pequeño, Alberto Comesaña-Campos and Alberto Fernández-Villar

Appl. Syst. Innov. 2025, 8(5), 128; https://doi.org/10.3390/asi8050128 - 2 Sep 2025

Viewed by 825

Abstract

Background: Chronic Obstructive Pulmonary Disease (COPD) often leads to acute exacerbations requiring hospitalization. While artificial intelligence (AI) has been increasingly used to improve COPD management, predicting whether the length of hospital stay (LOHS) will exceed clinically relevant thresholds remains insufficiently explored. Methods: This [...] Read more.

Background: Chronic Obstructive Pulmonary Disease (COPD) often leads to acute exacerbations requiring hospitalization. While artificial intelligence (AI) has been increasingly used to improve COPD management, predicting whether the length of hospital stay (LOHS) will exceed clinically relevant thresholds remains insufficiently explored. Methods: This study presents a novel clinical decision support system to predict whether LOHS following an acute exacerbation will surpass specific cutoffs (6 or 10 days). The approach involves two stages: (1) clinical, demographic, and social variables are encoded into structured signals and transformed into spectrogram-like images via a pipeline based on sinusoidal encoding and Mel-frequency cepstral coefficients (MFCCs); and (2) these images are fed into a Convolutional Neural Network (CNN) to estimate the probability of extended hospitalization. Feature selection with XGBoost reduced dimensionality to 16 variables. The model was trained and tested on a dataset of over 500 patients. Results: On the test set, the model achieved an AUC of 0.77 for predicting stays longer than 6 days and 0.75 for stays over 10 days. Sensitivity and specificity were 0.79/0.72 and 0.74/0.80, respectively. Conclusions: This system leverages image-based data formalization to predict LOHS in COPD patients, facilitating early risk stratification and more informed clinical planning. Results are promising, but external validation with larger and more diverse datasets remains essential. Full article

(This article belongs to the Special Issue Advancing Healthcare Through Intelligent Clinical Decision Support Systems: Techniques, Applications, and Future Directions)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Appl. Syst. Innov., Volume 8, Issue 5 (October 2025) – 40 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI