38139 research outputs found
Sort by
The social media ban is coming, whether families like it or not: 5 ways to prepare kids and teens
No full text31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '25)
No full textDespite the advancements in quantum convolution or quanvolution, challenges persist in making quanvolution scalable, efficient, and applicable to multi-dimensional data. Existing quanvolutional networks heavily rely on classical layers, with minimal quantum involvement due to inherent limitations in current quanvolution algorithms. Moreover, the application of quanvolution in the domain of 1D data remains largely unexplored. To address these limitations, we propose a new quanvolution algorithm-Quanv1D-capable of processing arbitrary-channel 1D data, handling variable kernel sizes, and generating a customizable number of feature maps, along with a classification network-fully quanvolutional network (FQN)-built solely using Quanv1D layers. Quanv1D is inspired by the classical Conv1D and stands out from the quanvolution literature by being fully trainable, modular, and freely scalable with a self-regularizing feature. To evaluate FQN, we tested it on 20 UEA and UCR time series datasets, both univariate and multivariate, and benchmarked its performance against state-of-the-art convolutional models (both quantum and classical). We found FQN to outperform all compared models in terms of average accuracy while using significantly fewer parameters. Additionally, to assess the viability of FQN on real hardware, we conducted a shot-based analysis across all the datasets to simulate statistical quantum noise and found our model robust and equally efficient
Heteronuclear Dual‐Atom Anchored g‐C3N4: p‐d Orbital Coupling Enable Efficient Urea Electrosynthesis from Gaseous Pollutants
No full textElectrocatalytic C–N coupling using gaseous pollutants NO and CO offers a promising alternative to conventional industrial urea synthesis. However, designing efficient electrocatalysts remains challenging due to the complexity of multi-step reactions, which yield diverse products. Herein, based on density functional theory (DFT) calculations, we explore Cu and p-block atoms (B, Al, and Ga) anchored on graphitic carbon nitride as novel heteronuclear double-atom catalysts (DACs) for urea synthesis from NO and CO. The reactants are stably adsorbed on the DACs, while strong d–p orbital hybridization facilitates effective activation and efficient C–N coupling. Among the candidates, CuB@g-C3N4 and CuGa@g-C3N4 exhibit particularly promising performance, with limiting potentials of −0.55 V and −0.36 V, respectively. Furthermore, these catalysts significantly suppress competing reactions, including the hydrogen evolution reaction (HER) and the formation of *NOH, *COH, and *CHO intermediates, ensuring high selectivity. Our work not only highlights highly efficient p-d DACs for electrocatalytic urea production but also provides a theoretical framework in catalyst design
Dynamic Control of Isolated Network Microgrids: A Resilient Backpropagation Neural Network-Based Virtual Inertia Control Approach
No full textIn isolated networked microgrid with high penetration of solar and wind-based generation, maintaining system stability and achieving optimal dynamic performance poses significant challenges due to reduced mechanical inertia traditionally provided by synchronous generators. This paper introduces a novel Resilient Back Propagation Bayesian Neural network-based virtual inertia control strategy to enhance frequency response and overall stability of the considered network microgrid. By leveraging robust control techniques, the proposed approach provides virtual inertia to respond effectively to varying system conditions and disturbances, improving system robustness and minimising the isolated networked microgrid’s tie-line power and frequency deviations. Comprehensive simulations, including case studies under varying disturbance conditions such as load fluctuations and renewable energy variations, prove the efficacy of the introduced virtual inertia control strategy. The proposed method outperforms conventional proportional integral derivative, and artificial bee colony-optimised proportional integral derivative control techniques in key dynamic performance metrics. It achieves the lowest integral time absolute error of 18.3912, compared to 30.8946 for proportional integral derivative and 20.8212 for optimised proportional integral derivative control techniques, demonstrating superior frequency response. Additionally, it achieves a mean square error of 4.0897e-07, significantly lower than 4.239e-06 for neural network-based fractional order proportional integral derivative control and 3.072e-06 for feed-forward neural network, confirming improved accuracy. Results indicate improved dynamic performance metrics, including faster frequency stabilization and reduced overshoot with minimal tie-line power and frequency deviation, compared to proportional integral derivative and optimal control techniques. The strategy’s adaptability and computational efficiency offer practical ..
Adversarial training-based robust model for transmission line’s insulator defect classification against cyber-attacks
No full textThe increased sophistication of smart grids has generated significant interest in employing unmanned aerial vehicles (UAVs) to monitor the operational condition of insulators, especially in identifying insulator defects to avoid substantial power loss, shortened lifespan of power lines, and power outages. The classification process of insulator defects remains a concern due to factors such as the small size of the objects, intricate backgrounds, and a restricted amount of available data. Moreover, although cyber-attacks can affect deep learning (DL) models to cause misclassification of insulator defects, existing literature has yet to address this issue. Hence, this work employs the YOLOv9 model, a state-of-the-art object detector, to classify defects in power transmission line insulators, including accurate identification of small defects within intricate backgrounds. Additionally, this study is the first to introduce fast gradient sign method (FGSM) and projected gradient descent (PGD)-based adversarial attacks to insulator defect classification tasks to address their severity, while proposing adversarial training as a defensive measure. The experimental results show that the YOLOv9 model achieves a mean average precision (mAP) value of 96.5%, outperforming existing YOLOv8s, YOLOv7, YOLOv5s, and Faster R-CNN by 1.4%, 1.8%, 6.7%, and 10.3%, respectively. Also, this investigation demonstrates the severe effects of adversarial attacks, where mAP values of the YOLOv9 model decrease up to 14% and 9.9% under FGSM and PGD attacks, respectively. The proposed adversarial trained YOLOv9 model ensures robustness against cyber-attacks and maintains high accuracy in normal conditions, even in the presence of Gaussian noise
Truss-inspired ultra-high strength, fire-safe, and thermal insulating double-crosslinked wood aerogels
No full textBio-based wood aerogel is one of the most promising materials to replace traditional petrochemical-based insulation materials. However, the flammability and poor mechanical strength of bio-based wood aerogels limit their applications in emerging fields. Inspired by a truss-supporting system, this study prepared a multifunctional bio-based cross-linked wood aerogel (TSP@Ca) by a dual hydrogen-ionic bonding strategy involving an oxidized wood cellulose framework, sodium alginate, phytic acid (PA), and Ca2+. Finite element simulation and mechanical analysis indicated that the multi-point support structure, resembling a truss framework, formed in the oxidized wood template significantly improved the strength of TSP@Ca aerogel (9.99 MPa), with a 154.84 % enhancement relative to that of oxidized delignified wood (TODW). The limiting oxygen index of TSP@Ca3 aerogel was as high as 43.3 %, and it can extinguish immediately when the fire was removed. The introduction of PA and Ca2+ promoted the dehydration, cross-linking, and charring of TSP@Ca aerogel, while the produced phosphorus-containing free radicals played an inhibitory role in the gas phase. Therefore, the peak of heat release rate of TSP@Ca aerogel was 80.66 % lower than that of TODW, showing excellent fire safety. Benefiting from the complex heat conduction path and enhanced interface resistance, the thermal conductivity of TSP@Ca was 46.4 % lower than that of TODW. The resulting aerogel combines ultra-high mechanical strength, excellent fire resistance, and thermal insulation, aligning with “green” development goals and offering broad application potential in construction, rail transport, and new energy sectors
Enabling 'Successful Knowledge Transfer' in Early Childhood Settings – A Meta Strategic Leadership Approach
No full textThe increase in knowledge access and professional development (PD) uptake by early childhood educators amplifies the need for innovative strategies to support successful knowledge transfer (SKT) in Early childhood Education and Care (ECEC) settings. SKT refers to the effective application of key learnings from PD events into practice. Existing literature highlights various factors affecting the successful implementation of PD learnings after the event with some pointing to the lack of fit for purpose leadership as one of the barriers. A study was conducted among educators in excellent rated centres in Australia to identify how they successfully implemented PD learnings following PD attendance. Data was collected using open ended interviews, informal observations and document analysis methods. The findings revealed that meta strategic leadership was a key enabler of SKT. This article discusses how meta-strategic leadership enables SKT within early childhood settings and the implications for practice
Committee Diversity Effect on Corporate Investment Risk Practices
No full textBackground: This study examines how diversifying committees influence corporate investment risk practices, specifically in decision‑making and resource allocation strategies. Previously, board diversity was commonly used in studies, but committee diversity was often overlooked, even though committees are delegated with providing recommendations for board decisions. Methods: Using information on committee presence, size, gender representation, and independent and non‑executive members, we build a detailed diversity composite index. We capture this information from various sources such as corporate official disclosures, corporate websites, and other relevant disclosures. We combine this data with financial and investment information collected through secondary data, including Bloomberg and Refinitiv databases about companies listed on the ASX 300 in the Australian equity market from 2018 to 2020. Results: Our findings show that diversity plays a much more critical role in enhancing long‑term strategic investment decisions than in driving short‑term operational gains. Conclusions: Additional investigations have shown that increased diversity enhances corporate resource allocation, generating optimal investment and investment efficiency levels. These findings highlight the strategic importance of diversity as a contributor to good governance and better financial performance
