67639 research outputs found
Sort by
Importance sampling for rare event tracking within the ensemble Kalman filtering framework
No full textIn this work we employ importance sampling (IS) techniques to track a small over-threshold probability of a running maximum associated with the solution of a stochastic differential equation (SDE) within the framework of ensemble Kalman filtering (EnKF). The proposed method acts as a post-processing step applied to the EnKF output: it uses the ensemble at a given observation time to estimate the probability of a rare event occurring before the next observation, without altering the EnKF itself. Between two observation times of the EnKF, we propose to use IS with respect to the initial condition of the SDE, IS with respect to the Wiener process via a stochastic optimal control formulation, and combined IS with respect to both initial condition and Wiener process. Both IS strategies require the approximation of the solution of Kolmogorov Backward equation (KBE) with boundary conditions. In multidimensional settings, we employ a Markovian projection dimension reduction technique to obtain an approximation of the solution of the KBE by just solving a one dimensional PDE. The proposed ideas are tested on three illustrative examples: Double Well SDE, Langevin dynamics and noisy Charney-deVore model, and showcase a significant variance reduction compared to the standard Monte Carlo method and another sampling-based IS technique, namely, multilevel cross entropy.Open access publishing provided by King Abdullah University of Science and Technology (KAUST). This work was supported by the KAUST Office of Sponsored Research (OSR) under Award No. URF/1/2584-01-01 and the Alexander von Humboldt Foundation. E. von Schwerin, G. Shaimerdenova and R. Tempone are members of the KAUST SRI Center for Uncertainty Quantification in Computational Science and Engineering
Burning characteristics of ethanol droplet suspended on NiCr wire with applied AC electric field
No full textThe effect of applied AC electric field on the burning behavior of ethanol droplets suspended on a NiCr wire (0.5 mm in diameter) was investigated by varying the AC frequency fAC (10–1000 Hz) and voltage VAC (1–7 kV). In the baseline case without applying electric field, internal recirculation was developed by heat transfer through the wire to the droplet due to Marangoni convection, enhancing the evaporation rate. Depending on VAC and fAC, three regimes can be identified: vertical oscillation of the droplet due to the combined effects of vertical electrostatic and dielectrophoretic forces (Regime I), combined oscillation and dripping, leading to electrospray of fine droplets from the surface by electrostatic force (Regime II); flame extinction due to substantial fuel loss via electrospray (Regime III). Ethanol droplets exhibited distinct dynamic behaviors from previously studied burning n-decane droplets, primarily due to the differences in fuel permittivity (affecting dielectrophoretic force) and electric conductivity (influencing dielectric relaxation frequency), both of which were not considered previously. Force-based scaling analysis revealed differences in oscillation amplitude and droplet response, and captured the onset conditions for dripping and extinction with fAC,cr∼VAC−p, in good agreement with experiments. The normalized droplet lifetime correlated strongly with key physical parameters, including the radial electric field gradient, AC frequency, and flame width and height.This work was supported by Basic Science Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (RS-2023–002451310 and RS-2024–00356149)
Concentrated solar power (CSP) driven desalination systems: A techno-economic review
No full textThe rising global demand for freshwater, coupled with the urgency to transition away from fossil fuel-based energy systems, has intensified research into sustainable desalination solutions. However, conventional desalination methods reliant on fossil fuels are highly energy-intensive, presenting substantial obstacles to achieving a low-carbon energy transition. Concentrated solar power (CSP) presents a compelling alternative, particularly for arid regions with high direct normal irradiation (DNI). This review provides a comprehensive analysis of recent advancements in CSP-driven desalination technologies, with a particular focus on key methods such as multi-stage flash distillation (MSF), multi-effect distillation (MED), membrane distillation (MD), and innovative hybrid systems. It systematically categorizes solar desalination technologies based on their functional components, economic feasibility, and research progress, highlighting advancements in hybrid system designs, thermal performance optimization, and economic evaluations. Although CSP desalination has experienced significant growth over the past five years, challenges remain in developing cost-competitive solutions, particularly in addressing parasitic losses during integration with conventional power systems. This review identifies potential strategies to overcome these challenges, including optimized system configurations, the integration of thermal energy storage, the adoption of advanced power cycles, and the hybridization of MED-RO systems. Realizing the full potential of CSP for sustainable freshwater production will require advances in materials, system integration, and hybrid configurations. A multidisciplinary approach—combining thermal sciences, desalination engineering, power systems, and techno-economic analysis, alongside supportive policies—is key to establishing CSP desalination as a viable solution for high-DNI, water-scarce regions. This review provides a timely and comprehensive overview of current progress and future directions, offering practical insights for advancing sustainable desalination technologies
Temporal Model-Based Federated Active Medical Image Classification
No full textTraditional federated learning relies on fully labeled datasets in each medical institution, which is impractical in real-world clinical scenarios. Federated Active Learning (FAL) addresses this by selecting a few informative samples for labeling, but it faces challenges such as domain shift across institutions. Besides, existing FAL methods rely on single-round model knowledge to estimate prediction-level uncertainty, ignoring uncertainty from features and model evolution during training. In this work, we propose TM-FAL, a novel framework for federated active medical image classification under domain shift. TM-FAL proposes a new uncertainty by integrating feature differences and prediction confidence from temporal local and global models to capture both local-global differences and the inherent complexity of images. Additionally, we use the prediction of the global model as pseudo labels to group images to mitigate class imbalance caused by uncertainty-based selection. Experiments on two medical image classification datasets demonstrate that TM-FAL outperforms various state-of-the-art methods.This work is supported by the Guangdong Science and Technology Department (No. 2024ZDZX2004), and Guangdong Provincial Key Lab of Integrated Communication, Sensing and Computation for Ubiquitous Internet of Things(No. 2023B1212010007)
An experimental study on combustion behavior of candle flames in hypergravity
No full textThis work explores the combustion characteristics of candle flames with various wick diameters and lengths for the gravity level of 3 − 9 g by utilizing a centrifuge. Results show that the burning rate and flame height decrease with increasing gravity, as the suppression of capillary action inside the wick reduces the fuel supply. There exists a critical gravity level Gcr to determine whether the liquid wax can reach the wick tip. When G > Gcr at hypergravity, the liquefied wax cannot reach the wick tip, resulting in reduced flame height and burning rate, whereas at G < Gcr, sufficient liquid wax is supplied to generate higher flame height. The candle flame has a transition from a stable laminar to oscillating flame, because of the enhanced buoyancy-induced velocity in hypergravity. However, a flame extinction is observed at 9 g due to minimized fuel supply rate. Considering the capillary-driven fuel supply mechanism and enhanced buoyant flow, flame oscillation frequency and amplitude are found to initially increase and then decrease with two oscillation modes (bulk flickering and tip flickering). The oscillation frequencies can well be described in terms of the Strouhal and Froude number relationship incorporating the change of burning rate. A physical model of burning rate is established considering the variation of characteristic length. The flame height in hypergravity is well predicted based on the Roper's model by adopting the "apparent port burner" concept, taking into account the effect of fuel supply rate. This paper provides comprehensive experimental data and facilitates fundamental understanding regarding the candle flames in hypergravity.This research was funded jointly by National Natural Science Foundation of China (Nos. 52225605, 52306171), Space Application System of China Manned Space Program under the funded project (KJZ-YY-NRS0602), China National Postdoctoral Program for Innovative Talents (BX20230354) and New Cornerstone Science Foundation through the XPLORER PRIZE
Unlocking mid-to-long-term flexibility: why seasonal pumped storage outperforms conventional pumped storage in wind-solar dominated grids
No full textGlobal efforts toward Dual Carbon Goals have spurred rapid growth in wind and solar installations worldwide. However, the inherent randomness and intermittency of wind and solar pose critical challenges to long-term flexibility requirements in power systems. While pumped storage remains a crucial regulating power source, the differential effectiveness between seasonal pumped storage (SPS) and conventional pumped storage (CPS) in mitigating medium-to-long-term renewable fluctuations has not been systematically investigated. Therefore, this study provides a comprehensive evaluation of the performance of SPS and CPS in addressing stochastic fluctuations of wind and solar. First, the output characteristics of wind and solar resources were analyzed, and multiple wind and solar scenarios were developed to simulate the regulatory differences between SPS and CPS under varying wind and solar scenarios. Subsequently, this study proposes an integrated evaluation framework combining entropy-weighted and coefficient of variation methods to objectively assess the technical, economic, and stability indicators, using Qinghai Province as a case study to analyze the advantages of both SPS and CPS. The results show that SPS outperforms CPS across many technical indicators. Notably, the carbon emission reduction effect of SPS becomes more pronounced in scenarios with higher wind power capacity. Economically, it has to be admitted that the investment cost of SPS is 1.33 times that of CPS. Although the operating cost is one fifth of CPS, the levelized cost of electricity of SPS is still higher than CPS. Finally, the comprehensive evaluation demonstrates the superiority of SPS, achieving an overall benefit score of 98.03, in stark contrast to 68.6 for CPS. These findings offer critical insights for energy planners and policymakers in optimizing storage solutions to enhance grid flexibility and accelerate decarbonization.This work was supported by the following grants: National Natural Science Foundation of China (Grant No.: 52409120), National Natural Science Foundation of China (Grant No.: 52339006), Free Exploration Basic Research (Grant No.: 2024ZY-JCYJ-02-37), China Postdoctoral Science Foundation, 75th Batch of General Projects (Grant No.: 2024M752626), Postdoctoral Fellowship Program of CPSF (Grant No.: GZC20232157), and Shaanxi Province Postdoctoral Research Project (Grant No.: 2023BSHEDZZ105)
Effect of Donors on Intramolecular Photoinduced Charge-Transfer in π-Conjugated Donor-Acceptor Solutions for Solar Cells Applications.
No full textPhotoinduced charge transfer kinetics are essential in assessing the appropriateness of organic materials for solar cell applications. This study utilized femtosecond-to-microsecond time-resolved spectroscopy to examine the excited-state kinetics of a series of π-conjugated donor–acceptor polymers featuring the potent electron-withdrawing unit ((E)-1,2-di(thiazol-2-yl) diazene) (ATZ) alongside three distinct donor moieties: thieno [3,2-b] thiophene (TT), 2,2′-bithiophene (BT), and (E)-1,2-di(thiophen-2-yl) ethene (DTV). Steady-state absorption, fluorescence, density functional theory (DFT) calculations, and transient absorption (TA) spectroscopy were conducted to clarify their photophysical behavior in the presence of 1,4-dicyanobenzene (DCB), a recognized electron acceptor. The findings indicate that intramolecular charge transfer from the donor segments (TT, BT, and DTV) to DCB predominates the excited-state deactivation mechanisms. DFT studies reveal that the highest occupied molecular orbitals (HOMOs) are delocalized along the polymer backbone, whereas the lowest unoccupied molecular orbitals (LUMOs) are predominantly localized on the donor units, hence promoting efficient electron transfer to DCB. These findings elucidate the charge-transfer mechanisms of ATz-based donor–acceptor polymers and underscore their potential for solar applications.The author gratefully acknowledges the funding of the Deanship of Graduate Studies and Scientific Research, Jazan University, Saudi Arabia, through project number: (RG24-S060), and King Abdullah University of Science and Technology (KAUST) for facilitating the lab work
Survey of Load-Altering Attacks Against Power Grids: Attack Impact, Detection and Mitigation
No full textThe growing penetration of IoT devices in power grids despite its benefits, raises cybersecurity concerns. In particular, load-altering attacks (LAAs) targeting high-wattage IoT-controllable load devices pose serious risks to grid stability and disrupt electricity markets. This paper provides a comprehensive review of LAAs, highlighting the threat model, analyzing their impact on transmission and distribution networks, and the electricity market dynamics. We also review the detection and localization schemes for LAAs that employ either model-based or data-driven approaches, with some hybrid methods combining the strengths of both. Additionally, mitigation techniques are examined, focusing on both preventive measures, designed to thwart attack execution, and reactive methods, which aim to optimize responses to ongoing attacks. We look into the application of each study and highlight potential streams for future research.This work has been supported in part by PhD Cofund WALL-EE project between the University of Warwick, UK and CY Cergy Paris University, France, EPSRC under Grant EP/S035362/1, and King Abdullah University of Science and Technology (KAUST) under Awards No. ORFS-2022-CRG11-5021 and No. RFS-OFP2023-5505
Flexible Temperature Sensor using Laser Induced Graphene (LIG) based on Processing-Controlled Seebeck Contrast
No full textFlexible temperature sensors and thermocouples offer precise temperature measurement on curved and irregular surfaces, making them ideal for wearable, environmental, and structural monitoring applications. This work presents a flexible Seebeck effect-based temperature sensor fabricated using Laser-Induced Graphene (LIG). Unlike conventional thermocouples, each leg was created with different laser configurations, where the defective structure of LIG plays a crucial role in enhancing the Seebeck coefficient by increasing phonon scattering at grain boundaries. Higher laser fluence leads to an increased IG/ID ratio, indicating a higher defect density, and a reduction in crystalline size (La), which results in more grain boundaries that further improve thermoelectric performance. To enhance sensitivity, a thermopile approach was implemented, connecting two LIG thermocouples in series, effectively increasing the Seebeck coefficient from 2.68 µV/oC to 5.67 µV/oC. The proposed sensor demonstrates great potential for next-generation temperature-sensing applications in various fields, including flexible electronics, healthcare, and industrial monitoring
Wafer-scale radio frequency ZnO Schottky diodes and arithmetic circuits
No full textModern telecommunication technologies, such as the 5G and upcoming 6G networks, rely on devices operating in the radio frequency (RF) spectrum of 0.3–90 GHz and 7–300 GHz, respectively. To meet these demanding frequency requirements, new manufacturing methods and device architectures are gaining increasing attention. However, achieving scalable manufacturing alongside ultra-fast device operation presents formidable techno-economic challenges. Here, we explored a modified version of adhesion lithography (a-Lith) to create coplanar nanogap zinc oxide (ZnO) Schottky diodes for application in diode-logic arithmetic circuits. The planar ZnO diodes offer highly scalable manufacturing and combine high current rectification (> 106) with low reverse currents (≈80 pA) and a remarkable cut-off frequency of over 25 GHz. Engineering the topologies of the planar ZnO diodes enables their facile monolithic integration into multi-bit AND and OR gates over 4-inch glass wafers. By integrating several such logic gates, we demonstrated fully functional monolithic 2-bit Half-Adder circuits, the primary component of an arithmetic logic unit. The work offers an alternative method for developing fast large-area electronics that could lead to a new family of logic circuitry.King Abdullah University of Science and Technology,BAS/1/1389-01-01