A5082218019- Hydrology and Sediment Transport Processes
- Graphene research and applications
- Wind Energy Research and Development
- Model Reduction and Neural Networks
- Flood Risk Assessment and Management
- Hydrology and Watershed Management Studies
- Adaptive optics and wavefront sensing
- Fluid Dynamics and Vibration Analysis
- Nanowire Synthesis and Applications
- Wind and Air Flow Studies
- Optical Polarization and Ellipsometry
- Optical Systems and Laser Technology
- Advanced optical system design
- Infection Control and Ventilation
- Aerodynamics and Acoustics in Jet Flows
- Aerodynamics and Fluid Dynamics Research
- Advanced Photocatalysis Techniques
- Supercapacitor Materials and Fabrication
- Carbon Nanotubes in Composites
- Silicon and Solar Cell Technologies
- Lattice Boltzmann Simulation Studies
- Electrocatalysts for Energy Conversion
- Energy Load and Power Forecasting
- Advanced Optical Imaging Technologies
- Electrical and Bioimpedance Tomography
Stony Brook University
2020-2024
Beijing Institute of Technology
2018-2024
China University of Petroleum, East China
2021-2022
Tsinghua University
2013-2018
State Key Laboratory of New Ceramics and Fine Processing
2016-2018
Xinjiang Normal University
2018
Xi'an Jiaotong University
2017
Nanjing Foreign Language School
2015
Materials Science & Engineering
2014
Graphene (G) films were grown on copper foils by chemical vapor deposition and transferred onto n-type silicon (Si) to form G/Si Schottky heterojunction solar cells. The power conversion efficiencies (PCEs) of the cells in range 1.94–2.66%. Four volatile oxidants HNO3, HCl, H2O2 SOCl2 employed treat graphene cells, PCEs could be greatly enhanced after being treated all doping showed best improvement. A cell with an initial PCE 2.45% increased 5.95% upon treatment. stability oxidant-treated...
The Coronavirus disease outbreak of 2019 has been causing significant loss life and unprecedented economical throughout the world. Social distancing face masks are widely recommended around globe in order to protect others prevent spread virus through breathing, coughing, sneezing. To expand scientific underpinnings such recommendations, we carry out high-fidelity computational fluid dynamics simulations resolution realism elucidate underlying physics saliva particulate transport during...
In meandering rivers, interactions between flow, sediment transport, and bed topography affect diverse processes, including bedform development channel migration. Predicting how these the spatial patterns magnitudes of deformation in rivers is essential for various river engineering geoscience problems. Computational fluid dynamics simulations can predict morphodynamics at fine temporal scales but have traditionally been challenged by large scale natural rivers. We conducted coupled...
MoS2-based electrocatalysts are promising cost-effective replacements for Pt-based catalysts hydrogen evolution by water splitting, yet achieving high current density at low overpotential remains a challenge. Herein, binder-free electrode of MoS2/CNF (carbon nanofiber) is prepared electrospinning and subsequent thermal treatment. The growth MoS2 nanoplates contained within or protruding out from the CNF can be controlled adding urea ammonium bicarbonate to precursors, due cross-linking...
Anti-reflective transparent electrodes are achieved demonstrating a highest efficiency of 13% and short-circuit current density 34 mA cm<sup>−2</sup>.
A convolutional neural network (CNN) autoencoder model has been developed to generate 3D realizations of time-averaged velocity in the wake wind turbines at Sandia National Laboratories Scaled Wind Farm Technology (SWiFT) facility. Large-eddy simulations (LES) SWiFT site are conducted using an actuator surface simulate turbine structures produce training and validation datasets CNN. The validated SpinnerLidar measurements wakes instantaneous fields from LES used train trained CNN is then...
Abstract Prediction of statistical properties the turbulent flow in large‐scale rivers is essential for river analysis. The large‐eddy simulation (LES) provides a powerful tool such predictions; however, it requires very long sampling time and demands significant computing power to calculate turbulence statistics riverine flows. In this study, we developed encoder‐decoder convolutional neural networks (CNNs) predict first‐ second‐order meandering using instantaneous LES results. We train...
Large-area (e.g. centimeter size) graphene sheets are usually synthesized via pyrolysis of gaseous carbon precursors methane) on metal substrates like Cu using chemical vapor deposition (CVD), but the presence grain boundaries and residual polymers during transfer deteriorates significantly properties CVD graphene. If nanotubes (CNTs) can be covalently bonded to graphene, hybrid system could possess excellent electrical conductivity, transparency mechanical strength. In this work, conducting...
A multifunctional device combining photovoltaic conversion and toxic gas sensitivity is reported. In this device, carbon nanotube (CNT) membranes are used to cover onto silicon nanowire (SiNW) arrays form heterojunction. The porous structure large specific surface area in the heterojunction both benefits for adsorption. virtue of these merits, doping a feasible method improve cell's performance can also work as self-powered sensor beyond solar cell. It shows significant improvement cell...
We present a physics-enhanced convolutional neural network (PECNN) algorithm for reconstructing the mean flow and turbulence statistics in wake of marine hydrokinetic (MHK) turbine arrays installed large-scale meandering rivers. The embeds mass momentum conservation equations into loss function PECNN to improve physical realism reconstructed fields. is trained using large eddy simulation (LES) results single row turbines virtual river. Subsequently, applied predict MHK with arrangements...
This study proposes a novel machine learning (ML) methodology for the efficient and cost-effective prediction of high-fidelity three-dimensional velocity fields in wake utility-scale turbines. The model consists an autoencoder convolutional neural network with U-Net skipped connections, fine-tuned using data from large-eddy simulations (LES). trained takes low-fidelity field cost-effectively generated analytical engineering as input produces fields. accuracy proposed ML is demonstrated wind...
Four different graphene films were synthesized via chemical vapor deposition by using acetonitrile with feed rates of 0.01, 0.02, 0.04, and 0.06 mL/min. Heterojunction solar cells assembled transferring as‐synthesized onto n ‐Si. Solar based on samples produced at mL/min demonstrate power conversion efficiencies 2.26%, 2.10%, 1.02%, 0.94%, respectively. When HNO 3 was used to dope the films, corresponding photovoltaic increased 4.98%, 4.19%, 2.04%, 1.74%, Mechanism for improved efficiency...
Cophasing is crucial for segmented or sparse aperture telescopes to achieve high resolution. In this Letter, we propose a novel, the best of our knowledge, model-based piston correction method that can remove large-scale errors within few iterations using extended objects. The relation between error and metric function derived theoretically under broadband illumination. based on image's power spectral density at spatial frequency where sidelobe peak modulation transfer (MTF) appears....
The challenge in developing a protection barrier from corrosion is to develop an economical yet efficient method for the coating material. Graphene has been deemed be ideal material anti-corrosion, its long-term performance disappointing due galvanic corrosion. In effort address issues associated with graphene materials, we have developed amorphous nitrogen doped carbon (a-CN) film exhibiting remarkable anti-corrosion properties. of this appears superior because strong ion transportation...
This study proposes a cost-effective machine-learning based model for predicting velocity and turbulence kinetic energy fields in the wake of wind turbines yaw control applications. The consists an auto-encoder convolutional neural network (ACNN) trained to extract features turbine wakes using instantaneous data from large-eddy simulation (LES). proposed framework is demonstrated by applying it Sandia National Laboratory Scaled Wind Farm Technology facility consisting three 225 kW turbines....
This Letter proposes a snapshot imaging spectrometer, which obtains the spectral information and spatial in one “shot”. The device proposed can achieve data cube size of 21×29×40 waveband 400–800 nm. core element this system is microlens array, contains 60×60 microlenses square arrangement, each has an aperture 125 μm×125 μm, F number 15. array mounted rotation mount, provides 360° around optical axis to maximize resolution. final resolution about 10