Saved Queries

The deployment of Discrete Variable Quantum Key Distribution (DV-QKD) in high-traffic, short-reach environments, such as intra-data center networks, is currently constrained by the saturation of single-photon detectors. While CMOS Single-Photon Avalanche Diodes (SPADs) offer a cost-effective solution, their Secure Key Rate (SKR) is limited by detector dead time. To the best of the authors’ knowledge, this work is the first to derive a generalized detection rate model for SiPMs that addresses the dead-time bottlenecks of gigahertz-rate quantum cryptography. While methods for managing deadtime via active optical switching have been proposed, our model quantifies the benefits of passive spatial multiplexing inherent in standard SiPM arrays. Furthermore, contrasting with models designed to optimize energy resolution or characterize nonlinear charge response to light pulses, our work focuses on maximizing the detection count rate. We derive exact detection rate models for both analog (paralyzable) and digital (non-paralyzable) SiPM architectures, incorporating correlated noise sources such as optical crosstalk and afterpulsing. Simulation results indicate that SiPMs can increase detection rates by over an order of magnitude compared to single SPADs. Full article

(This article belongs to the Topic Responsible Classic/Quantum AI Technologies for Industrial Applications)

17 pages, 2678 KB

Open AccessArticle

Exploring the Role of TSPO-PET Imaging Among MRI-Negative Patients with Temporal Lobe Epilepsy: From the Perspective of Heterogeneity

by Yuncan Chen, Jing Wang, Shimin Xu, Qinyue Wang, Shuhao Mei, Jiaying Lu, Yiqiao Wang, Huamei Lin, Dongyan Wu, Liang Chen, Chuantao Zuo, Yihui Guan, Jingjie Ge and Xunyi Wu

Brain Sci. 2026, 16(2), 246; https://doi.org/10.3390/brainsci16020246 (registering DOI) - 22 Feb 2026

Abstract

Background/Objectives: This study explored the heterogeneous distribution pattern of translocator protein 18kDa (TSPO)-PET/MRI using radioligand [¹⁸F] DPA-714 in temporal lobe epilepsy patients and identified clinical factors influencing imaging outcomes. Methods: The TSPO imaging in individual patient was evaluated with both visual reading and quantitative assessment using an asymmetry index based on cerebellum-normalized standardized uptake values. The association between clinical factors and TSPO imaging outcomes was assessed. Pathological evaluation was conducted in three patients. Results: Twenty-nine TLE patients and ten healthy controls were enrolled. Visual evaluation revealed increased [¹⁸F] DPA-714 uptake in twenty patients as compared to controls, predominantly in a unilateral regional brain, while the remaining nine patients showed visually undetectable uptake of [¹⁸F] DPA-714. Consistently, quantitative analysis revealed that 69% (20/29) patients exhibited at least one brain area with significant asymmetry index, notably in the temporal lobe (85%, 17/20). A high asymmetry index could also be observed in the parietal (13.8%, 4/29) and occipital lobe (17.2%, 5/29). Significant associations were identified between the asymmetry index and seizure frequency (p = 0.045, OR = 7.994), and the interval from last seizure to PET scan (p = 0.033, OR = 6.712). Moreover, we confirmed the pathology in three patients via immunohistochemistry, which underscored the potential of TSPO-PET in detecting minor lesion. Conclusions: TSPO-PET reveals patient-specific and network-level neuroinflammatory heterogeneity in MRI-negative TLE, supporting its potential role as a complementary tool for presurgical evaluation. Full article

(This article belongs to the Section Neurotechnology and Neuroimaging)

►▼ Show Figures

Figure 1

19 pages, 3857 KB

Open AccessArticle

Aerodynamic Analysis and Design of a Sliding Drag Reduction System Using Graph Neural Networks

by Shinji Kajiwara and Cinto Ton

Fluids 2026, 11(2), 59; https://doi.org/10.3390/fluids11020059 (registering DOI) - 22 Feb 2026

Abstract

To maximize competitive performance in motorsports, balancing high downforce for cornering with low drag for straight–line speed is essential. This paper presents the development and optimization of a sliding Drag Reduction System (DRS) integrated with a ducktail guide for a Student Formula racing car. To overcome the computational costs and time constraints of conventional CFD–based iterative design, a Graph Neural Network (GNN) surrogate model was developed to predict aerodynamic coefficients. Unlike traditional models, the GNN directly learns from the geometric graph structure of the multi–element wing, enabling near–instantaneous and highly accurate predictions. CFD results indicated that activating the DRS reduced drag from 82.68 N to 25.51 N, improving the lift–to–drag ratio from 1.67 to 2.67. The GNN surrogate model achieved an R² value exceeding 0.99, demonstrating exceptional predictive fidelity compared to high–resolution simulations. Physical track testing with a Formula SAE vehicle corroborated these findings, showing a 4.6% improvement in 50 m acceleration and a 5.8% increase in maximum speed. This research establishes that GNN–based surrogate models can significantly accelerate the design and optimization of complex variable aerodynamic systems, providing a robust framework for performance enhancement in racing applications. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence in Fluid Mechanics, 2nd Edition)

►▼ Show Figures

Figure 1

21 pages, 635 KB

Open AccessArticle

A New Insight into Ancient Wheat Pasta: Physicochemical, Technological and Cooking Quality of Triticum dicoccum (Emmer)

by İzzet Özhamamcı

Appl. Sci. 2026, 16(4), 2138; https://doi.org/10.3390/app16042138 (registering DOI) - 22 Feb 2026

Abstract

Emmer (Triticum turgidum ssp. dicoccum) is attracting renewed interest as a nutrient-dense ancient wheat for sustainable cereal foods; however, product-level evidence for region-specific landraces remains limited. This study characterizes pasta produced exclusively from 100% Triticum dicoccum semolina cultivated in Ardahan (Türkiye) by integrating proximate composition, cooking performance, and instrumental texture (TPA). The emmer pasta contained 12.70% protein, 4.93% total dietary fiber, and 1.68% ash, with an energy value of 366.25 kcal/100 g. Cooking tests revealed 10.86% cooking loss, 219.98% water absorption, and 101.62% volume increase, indicating limited cooking tolerance consistent with a weaker starch–protein matrix. In comparison with conventional T. durum pasta, cooked emmer pasta exhibited comparable hardness, gumminess, and chewiness, but higher adhesiveness and springiness alongside lower resilience and cohesiveness. These results highlight Ardahan-grown T. dicoccum as a nutritionally valuable pasta raw material, albeit with technological constraints (particularly cooking loss) that warrant further optimization for industrial use. Full article

(This article belongs to the Section Food Science and Technology)

►▼ Show Figures

Figure 1

13 pages, 2014 KB

Open AccessArticle

Highly Thermally Conductive PDMS/h-BN Composites Enabled by Aspect-Ratio-Driven Alignment

by Mi-Ri An, Ji-Yoon Ahn, Eun-Taek Hor and Sung-Hoon Park

Polymers 2026, 18(4), 539; https://doi.org/10.3390/polym18040539 (registering DOI) - 22 Feb 2026

Abstract

Shear-induced alignment of hexagonal boron nitride (h-BN) platelets offers a scalable route to high-performance, electrically insulating thermal management materials, yet the role of filler geometry under practical shear processing remains unclear. Here, we examine how platelet aspect ratio governs alignment and heat transport in PDMS/h-BN composites processed by sequential roll-gap controlled two-roll milling. Using a geometric moment-arm perspective, we relate the platelet effective radius to the shear-driven rotational driving moment. High-aspect-ratio platelets (L-BN) exhibit more stable flow-parallel alignment than small platelets (S-BN), forming a better-connected conductive network. At 175 wt% loading, the aligned L-BN composite achieves 10.3 W m⁻¹ K⁻¹ (94% higher than its random counterpart) and outperforms the S-BN system while also improving stiffness and device-relevant heat dissipation. These results identify aspect ratio as an alignment-enabling design criterion for scalable thermal management. Full article

(This article belongs to the Special Issue Inorganic Filler-Based Polymeric Composites for Enhanced Mechanical and Thermal Properties)

►▼ Show Figures

Figure 1

25 pages, 19543 KB

Open AccessArticle

Enhancing Spatiotemporal Resolution of MCCA SMAP Soil Moisture Products over China: A Comparative Study of Machine Learning-Based Downscaling Approaches

by Zhuoer Ma, Peng Chen, Hao Chen, Hang Liu, Yuchen Zhang, Binyi Huang, Yang Hong and Shizheng Sun

Sensors 2026, 26(4), 1383; https://doi.org/10.3390/s26041383 (registering DOI) - 22 Feb 2026

Abstract

As a key parameter of the Earth’s ecosystem, soil moisture significantly influences land-atmosphere interactions and has important applications in meteorology, hydrology, and agricultural studies. However, existing passive microwave remote sensing products of soil moisture are limited by their discontinuous temporal coverage and relatively coarse spatial resolution (typically 25–55 km), which cannot meet the requirements for fine-scale applications. This study developed and compared four machine learning-based downscaling approaches to improve the spatiotemporal resolution of MCCA SMAP soil moisture products. The methodology involved establishing complex nonlinear relationships between soil moisture and various high-resolution surface parameters including albedo, evapotranspiration, precipitation, and soil properties. High-resolution soil moisture maps were generated by leveraging the scale-invariant characteristics between soil moisture and surface parameters, followed by comprehensive evaluation using in situ ground observations and triple collocation analysis. The results demonstrated that all downscaling models showed excellent consistency with original MCCA SMAP observations (R > 0.93, RMSE < 0.033 m³ m⁻³), while successfully providing enhanced spatial details. The Random Forest (RF) model exhibited superior performance, showing higher correlation coefficients and lower biases when compared with in situ measurements. Uncertainty analysis revealed relatively low uncertainty levels for all models except Backpropagation Neural Network (BPNN) model. The RF-downscaled products accurately tracked temporal variations of soil moisture and showed good responsiveness to precipitation patterns, demonstrating their potential for fine-scale hydrological applications and regional environmental monitoring. Full article

(This article belongs to the Section Environmental Sensing)

►▼ Show Figures

Figure 1

25 pages, 5640 KB

Open AccessArticle

Estimation of Winter Wheat SPAD Values by Integrating Spectral Feature Optimization and Machine Learning Algorithms

by Yufei Wang, Xuebing Wang, Jiang Sun, Zeyang Wen, Haoyong Wu, Lujie Xiao, Meichen Feng, Yu Zhao and Xianjie Gao

Agronomy 2026, 16(4), 489; https://doi.org/10.3390/agronomy16040489 (registering DOI) - 22 Feb 2026

Abstract

The chlorophyll content of plant leaves measured by the soil plant analysis development (SPAD) is an important indicator for measuring crop growth status and irrigation effect. The rapid, non-destructive and efficient estimation of crop SPAD values is of great significance to the field management of crops. In this study, the canopy hyperspectral reflectance and SPAD values of winter wheat were obtained, and the spectral curve was changed through four spectral processing methods, including first-order differential (FD), second-order differential (SD), multivariate scattering correction (MSC), and Savitzky–Golay smoothing (SG) to improve the correlation between canopy spectral reflectance and SPAD. Furthermore, to investigate and evaluate the performance of various vegetation indices (VIs) in estimating SPAD values for winter wheat, existing published indices were optimized using random band combinations derived from multiple canopy spectral transformations. The optimized vegetation index was used as the input variable of the model, and six machine learning algorithms, including random forest (RF), long short-term memory network (LSTM), multilayer perceptron (MLP), deep recurrent neural network (Deep-RNN), gated recurrent unit (GRU), and convolutional neural network (CNN), were used to construct the winter wheat SPAD values estimation model, and the model was verified. The experimental results demonstrate that, when utilizing an equivalent number of optimized vegetation indices as input, the GRU-based model achieves higher estimation accuracy compared to other models. Specifically, the coefficient of determination (R²) is improved by 0.12 compared to the RF model, by 0.03 compared to the LSTM model, by 0.12 compared to the MLP model, by 0.02 compared to the Deep-RNN model, and by 0.02 compared to the CNN model. At the same time, the GRU model also has a lower root mean square error (RMSE) and relative error (RE) of 7.37 and 24.90%, respectively. This study provides valuable hyperspectral remote sensing technology support for the implementation of winter wheat SPAD values estimation in the field. Full article

(This article belongs to the Section Precision and Digital Agriculture)

►▼ Show Figures

Figure 1

26 pages, 20085 KB

Open AccessArticle

Enhanced Sea Ice Classification Method for Dual-Polarization TOPSAR via Limited Full-Polarimetric Knowledge Distillation

by Di Yin, Jiande Zhang, Jiayuan Shen, Jitong Duan, Xiaochen Wang, Guangyao Zhou, Bing Han and Wen Hong

Remote Sens. 2026, 18(4), 666; https://doi.org/10.3390/rs18040666 (registering DOI) - 22 Feb 2026

Abstract

Accurate large-scale sea ice classification is vital for Arctic maritime activities. However, this task faces a fundamental challenge. Operational surveillance is restricted to wide-swath dual-polarization data, which limits classification precision due to polarimetric information deficiency. Conversely, while the quad-polarization mode offers the comprehensive scattering details required for more accurate classification, its narrow swath width prevents efficient large-scale monitoring. To address this challenge, we propose an enhanced sea ice classification method relying on limited co-region quad-polarization information to enhance dual-polarization data classification accuracy across larger spatiotemporal scales. Specifically, we construct a polarization-guided cross-mode knowledge distillation framework featuring an asymmetric teacher–student architecture. In this design, a hybrid CNN-Transformer teacher extracts robust scattering features from quad-polarization data to guide a lightweight student network operating on dual-polarization inputs. Through this transfer, the student acquires rich feature representations comparable to quad-polarization data, effectively compensating for the missing polarimetric scattering information. Experimental results on GF3-02 data demonstrate that the proposed method significantly outperforms the standalone dual-polarization network baseline, achieving an overall accuracy of 86.15%. This validates the effectiveness of the proposed method in enabling high-precision sea ice classification for large-scale monitoring. Full article

(This article belongs to the Special Issue Cutting-Edge PolSAR Imaging Applications and Techniques)

41 pages, 10740 KB

Open AccessArticle

Dynamic Multi-Relation Learning with Multi-Scale Hypergraph Transformer for Multi-Modal Traffic Forecasting

by Juan Chen and Meiqing Shan

Future Transp. 2026, 6(1), 51; https://doi.org/10.3390/futuretransp6010051 (registering DOI) - 22 Feb 2026

Abstract

Accurate multi-modal traffic demand forecasting is key to optimizing intelligent transportation systems (ITSs). To overcome the shortcomings of existing methods in capturing dynamic high-order correlations between heterogeneous spatial units and decoupling intra- and inter-mode dependencies at multiple time scales, this paper proposes a Dynamic Multi-Relation Learning with Multi-Scale Hypergraph Transformer method (MST-Hype Trans). The model integrates three novel modules. Firstly, the Multi-Scale Temporal Hypergraph Convolutional Network (MSTHCN) achieves collaborative decoupling and captures periodic and cross-modal temporal interactions of transportation demand at multiple granularities, such as time, day, and week, by constructing a multi-scale temporal hypergraph. Secondly, the Dynamic Multi-Relationship Spatial Hypergraph Network (DMRSHN) innovatively integrates geographic proximity, passenger flow similarity, and transportation connectivity to construct structural hyperedges and combines KNN and K-means algorithms to generate dynamic hyperedges, thereby accurately modeling the high-order spatial correlations of dynamic evolution between heterogeneous nodes. Finally, the Conditional Meta Attention Gated Fusion Network (CMAGFN), as a lightweight meta network, introduces a gate control mechanism based on multi-head cross-attention. It can dynamically generate node features based on real-time traffic context and adaptively calibrate the fusion weights of multi-source information, achieving optimal prediction decisions for scene perception. Experiments on three real-world datasets (NYC-Taxi, -Bike, and -Subway) demonstrate that MST-Hyper Trans achieves an average reduction of 7.6% in RMSE and 9.2% in MAE across all modes compared to the strongest baseline, while maintaining interpretability of spatiotemporal interactions. This study not only provides good model interpretability but also offers a reliable solution for multi-modal traffic collaborative management. Full article

(This article belongs to the Special Issue Recent Advances in Artificial Intelligence and Big Data for Intelligent Transportation Systems)

16 pages, 2520 KB

Open AccessArticle

Flow-Integrated Efficiency Assessment of Shared Bicycles and Its Influencing Factors: A Case Study of Beijing

by Zhifang Yin, Yiqi Li, Shengyao Qin and Teqi Dai

Appl. Sci. 2026, 16(4), 2137; https://doi.org/10.3390/app16042137 (registering DOI) - 22 Feb 2026

Abstract

As dockless bike-sharing systems rapidly expanded, this study aims to develop a flow-integrated framework for assessing bicycle usage efficiency, which addresses a critical gap in conventional static indicators. Existing studies rely primarily on big data to evaluate location-specific efficiency using Time-to-Booking (ToB). However, ToB ignores network flow effects while bicycles departing from the same location may reach destinations with vastly different ToB values. To overcome this, we propose a flow-integrated ToB (FwToB) index that incorporates the idle time at both the trip origin and destination. Applying this index to central Beijing reveals significant spatial heterogeneity while maintaining the original core-periphery pattern, indicating that most bicycles flow to areas with similar efficiency. Geographically weighted regression further shows that factors like population density, healthcare, shopping facilities, and distance to metro stations influence efficiency with substantial spatial non-stationarity. These findings advance the understanding of bike-sharing efficiency and offer insights for operators and urban planners. Full article

(This article belongs to the Section Earth Sciences)

►▼ Show Figures

Figure 1

3 pages, 149 KB

Open AccessEditorial

Special Issue “Inflammatory Airway Diseases: Diagnosis, Pathology, Molecular Mechanisms and Treatment Options”

by Emma Matthews and Stefanie Krick

Int. J. Mol. Sci. 2026, 27(4), 2057; https://doi.org/10.3390/ijms27042057 (registering DOI) - 22 Feb 2026

Abstract

The lung is a unique organ because of its continuous exposure to the external environment, which requires a complex and tightly regulated immune network [...] Full article

(This article belongs to the Special Issue Inflammatory Airway Diseases: Diagnosis, Pathology, Molecular Mechanisms and Treatment Options)

33 pages, 10643 KB

Open AccessArticle

Deciphering the Biosynthetic Pathways and Regulatory Networks of the Active Components of Cibotium barometz by Transcriptomic Analysis

by Yuli Zhang, Zhen Wang, Minghui Li, Ting Wang and Yingjuan Su

Int. J. Mol. Sci. 2026, 27(4), 2050; https://doi.org/10.3390/ijms27042050 (registering DOI) - 22 Feb 2026

Abstract

Cibotium barometz (L.) J. Sm., a medicinally significant fern in traditional Chinese medicine, is little explored at the genomic level regarding its bioactive compounds. Using an integrated approach combining Illumina and PacBio sequencing technologies, we profiled its root, rachis, and pinna transcriptomes, identifying 12,718, 21,341, and 11,441 unigenes, respectively. Our analysis systematically characterized the transcriptional features of transcription factors (TFs), simple sequence repeats (SSRs), long non-coding RNAs (lncRNAs), and differentially expressed genes (DEGs). Enrichment analyses highlighted the roles of highly expressed unigenes in secondary metabolism. Seventeen key enzymes involved in polysaccharide biosynthesis showed tissue-specific expression patterns. Notably, total polysaccharide content correlated positively with UDP-arabinose 4-epimerase (UXE) expression but negatively with phosphoglucomutase (PGM) and 3,5-epimerase/4-reductase (UER1). Flavonoid accumulation inversely correlated with chalcone synthase (CHS) expression. Two lignin pathways (H-lignin and G-lignin) were characterized, with phenylalanine ammonia-lyase (PAL), cinnamate-4-hydroxylase (C4H), and cinnamyl alcohol dehydrogenase (CAD) as key genes. The absence of ferulate-5-hydroxylase (F5H) explains the undetected S-lignin pathway. Regulatory network analysis revealed positive correlations between PAL expression and NAC72/NAC78/WRKY35 and C4H expression and WRKY65/WRKY69/WRKY71, while a negative correlation was revealed between flavonoid 3′,5′-hydroxylase (F3′5′H) and MYB3R4. This study provides comprehensive transcriptomic insights into C. barometz bioactive compound biosynthesis, serving as a foundation for mechanistic research. Full article

(This article belongs to the Section Molecular Genetics and Genomics)

►▼ Show Figures

Figure 1

50 pages, 1827 KB

Open AccessArticle

Shared Autoencoder-Based Unified Intrusion Detection Across Heterogeneous Datasets for Binary and Multi-Class Classification Using a Hybrid CNN–DNN Model

by Hesham Kamal and Maggie Mashaly

Mach. Learn. Knowl. Extr. 2026, 8(2), 53; https://doi.org/10.3390/make8020053 (registering DOI) - 22 Feb 2026

Abstract

As network environments become increasingly interconnected, ensuring robust cyber-security has become critical, particularly with the growing sophistication of modern cyber threats. Intrusion detection systems (IDSs) play a vital role in identifying and mitigating unauthorized or malicious activities; however, conventional machine learning-based IDSs often rely on handcrafted features and are limited in their ability to detect diverse attack types across disparate network domains. To address these limitations, this paper introduces a novel unified intrusion detection framework that implements “Structural Dualism” to integrate three heterogeneous benchmark datasets (CSE-CIC-IDS2018, NF-BoT-IoT-v2, and IoT-23) into a harmonized, protocol-agnostic representation. The framework employs a shared autoencoder architecture with dataset-specific projection layers to learn a unified latent manifold. This 15-dimensional space captures the underlying semantics of attack patterns (e.g., volumetric vs. signaling) across multiple domains, while dataset-specific decoders preserve reconstruction fidelity through alternating multi-domain training. To identify complex micro-signatures within this manifold, the framework utilizes a synergistic hybrid convolutional neural network–deep neural network (CNN–DNN) classifier, where the CNN extracts spatial latent patterns and the DNN performs global classification across twenty-five distinct classes. Class imbalance is addressed through resampling strategies such as adaptive synthetic sampling (ADASYN) and edited nearest neighbors (ENN). Experimental results demonstrate remarkable performance, achieving 99.76% accuracy for binary classification and 99.54% accuracy for multi-class classification on the merged dataset, with strong generalization confirmed on individual datasets. These findings indicate that the shared autoencoder-based CNN–DNN framework, through its unique feature alignment and spatial extraction capabilities, significantly strengthens intrusion detection across diverse and heterogeneous environments. Full article

23 pages, 2608 KB

Open AccessArticle

Designing Predictive Models: A Comparative Evaluation of Machine Learning Algorithms for Predicting Body Carcass Fat in Ewes at Weaning

by Ahmad Shalaldeh, Mosleh Abualhaj, Ahmad Adel Abu-Shareha, Ayman Elshenawy, Yassen Saoudi, Muzammil Hussain, Ahmad Shubita, Majeed Safa and Chris Logan

Agriculture 2026, 16(4), 488; https://doi.org/10.3390/agriculture16040488 (registering DOI) - 22 Feb 2026

Abstract

Accurate estimation of Body Carcass Fat (BCF) is essential for evaluating the physiological condition of ewes. Traditional assessment via Body Condition Score (BCS) through palpation is inaccurate and subjective. BCF can now be predicted more precisely using objective measurements. This study presents a comparative analysis of eight machine learning (ML) models for predicting BCF in Coopworth ewes, using weight and RGB-image-based body measurements. Four non-linear regression methods and four neural network architectures were evaluated using a dataset of 74 ewes with 13 independent variables. The dataset was partitioned into training (52 ewes), validation (11 ewes), and testing (11 ewes) sets. The Gradient Boosting Regression achieved the highest predictive accuracy with an R² value of 0.9434 using body weight and width, followed by Ensemble Neural Network (R² = 0.9371) using body weight. The findings demonstrate the effectiveness of the Gradient Boosting Regression, Ensemble Neural Network and Random Forest tree-based approaches for morphometric prediction tasks in biological applications. BCF values obtained from image analysis were validated against those derived from computerized tomography (CT), considered the gold standard. These findings highlight the potential of image-guided, ML-driven models for objective, non-invasive, cost-effective assessment of ewe body composition in modern livestock systems. Full article

(This article belongs to the Special Issue Machine Learning in Precision Livestock Farming: From Animal Activity Forecasting to Environmental Control)

►▼ Show Figures

Figure 1

22 pages, 3981 KB

Open AccessArticle

Rotating Electric Machine Fault Diagnosis with Magnetic Flux Measurement Using Deep Learning Models

by Obinna Onodugo, Innocent Enyekwe and Emmanuel Agamloh

Energies 2026, 19(4), 1106; https://doi.org/10.3390/en19041106 (registering DOI) - 22 Feb 2026

Abstract

This paper presents new techniques for electric machine diagnostics that combine advanced signal processing and artificial intelligence (AI)-based techniques using magnetic flux measurements acquired under various operating conditions. Developing an effective electric machine diagnostics tool is paramount for increased industrial productivity and extending the service life of the machine. The existing diagnostic tools face issues, including false indication of faults using classical methods, and the proposed data-driven methods based on machine learning lack transferability of model knowledge on an unseen dataset from different motor types or power ratings due to structural differences. To overcome these diagnostic drawbacks of statistical ML classifiers and classical approaches, innovative feature selection methods were employed in this work to preprocess the measured magnetic flux into a spectrogram image, and the transfer learning (TL) technique was applied to fine-tune convolution neural networks (CNNs) ImageNet pretrained models. The experimental results show the trained statistical ML classifiers and traditional CNN performance on unseen BU data and on the external data, and the performance demonstrated a lack of generalization on external datasets of different power ratings or structures. Models with such drawbacks cannot be used for developing effective diagnostic systems. The TL technique was employed on different deep CNN ImageNet pretrained models with spectrogram images as inputs to the deep CN network. This approach demonstrated an advanced and improved electric machine diagnostic system that addresses the drawbacks of the current ML-based diagnostic systems. The generalized model developed using CNN ResNet50 outperformed other deep CNN ImageNet models in correctly diagnosing faults on both the dataset generated from the authors’ lab and on an external dataset of a different machine from another research lab. Full article

(This article belongs to the Special Issue Innovations in Electric Motor Drives: Exploring the Future of High-Performance, Energy Efficiency and Reliability)

►▼ Show Figures

Figure 1

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 2000.

Go to page 1 2 3 4 5

Search Results (166,559)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI