A5021349170- Advanced Combustion Engine Technologies
- Combustion and flame dynamics
- Atmospheric chemistry and aerosols
- Heat transfer and supercritical fluids
- Structural Behavior of Reinforced Concrete
- Electrowetting and Microfluidic Technologies
- Microfluidic and Bio-sensing Technologies
- Seismic Performance and Analysis
- Handwritten Text Recognition Techniques
- Chemical Thermodynamics and Molecular Structure
- Structural Load-Bearing Analysis
- Model Reduction and Neural Networks
- Innovative Microfluidic and Catalytic Techniques Innovation
- Phase Equilibria and Thermodynamics
- User Authentication and Security Systems
- Powdery Mildew Fungal Diseases
- Hand Gesture Recognition Systems
- Fire effects on concrete materials
- Data Quality and Management
- Optical Wireless Communication Technologies
- Adaptive optics and wavefront sensing
- Orbital Angular Momentum in Optics
- Nuclear Engineering Thermal-Hydraulics
- Energetic Materials and Combustion
- Fire dynamics and safety research
Chongqing University
2022-2025
University of Chinese Academy of Sciences
2024
Tsinghua University
2020-2024
The University of Sydney
2024
Technical University of Munich
2022
UNSW Sydney
2022
Shanghai Jiao Tong University
2022
Duke University
2020-2021
Durham Technical Community College
2020
Advances in lab-on-a-chip technologies are driven by the pursuit of programmable microscale bioreactors or fluidic processors that mimic electronic functionality, scalability, and convenience. However, few mechanisms allow for basic logic operations on rewritable paths due to cross-contamination, which leads random interference between "fluidic bits" droplets. Here, we introduce a mechanism allows contact-free gating individual droplets based scalable features acoustic streaming vortices...
Decomposition and control of acoustic streaming enable contact-free manipulation bioanalytes digitalized liquid handling.
Advanced propulsion and power-generation systems often operate under extreme conditions, where thermophysical properties of the working fluids undergo complex variations in a wide range fluid states, empirical cubic equations state could yield substantial errors density prediction. The present work develops data-driven models for accurate estimation general across all thermodynamic regimes. model starts with equation state, whose alpha function is calibrated manner statistical correction...
Encoding information into the serial sequence of micro-droplet lengths<italic>via</italic>acoustofluidic dispensing.
A multiple neural network controller is proposed and demonstrated to suppress the pressure oscillation of Rijke tube acoustic network. This consists three modules including two separate networks, i.e., controlled object that pretrained before control trained in real time during process. can identify characteristics oscillating combustion, achieve adaptive output, extend applicability expansibility. Results show different stages using fuel valve or loudspeaker as actuators. When exact...
Recent years have witnessed a growing research interest in the rotational Doppler effect associated with orbital angular momentum of light, emerging as powerful tool to detect rotating bodies remote sensing. However, this method, when exposed turbulence realistic environment, has some severe limitations, leading unrecognizable signals overwhelmed background noise. Here we put forward concise yet efficient method that enables turbulence-resilient detection cylindrical vector beams....
Chemical kinetics plays an important role in the direct detonation initiation (DDI) of various combustible mixtures. However, its impact on dynamics has rarely been studied with detailed mechanisms. This study introduces active subspace method to systematically explore chemical unsteady DDI H2-O2-Ar mixture, a 13-species mechanism. The kinetic effects ZND structure sub-CJ shock speed (sub-CJ structure) is first investigated, where three reactions minimum speed, H + O2 = O OH, M OH and H2O...
The modeling of scalar mixing timescale remains a primary challenge in the transported probability density function (TPDF) method. variation among species, i.e., differential mixing, results from difference molecular diffusivity and reaction-induced gradient. Nevertheless, vast majority TPDF studies on turbulent non-premixed flames simply apply single determined by mixture fraction. In this work, (RIDM) model for individual species is proposed. key idea RIDM to approximate relative magnitude...
Modeling molecular mixing remains a key issue for the filtered density function (FDF) method. We propose new closure timescale of each individual chemical species. A posteriori tests demonstrate advantages proposed model, highlighting its potential to be employed in FDF simulations turbulent reacting flows.
Abstract Modal interference in a multimode fiber (MMF) has been utilized to develop simple sensors for various physical parameters. Herein, first, we investigate the dependences of core diameter and numerical aperture (NA) MMF on performance multimode-interference-based strain temperature sensing. We find that larger leads higher sensitivity but lower (absolute value) NA does not influence results value). Subsequently, using obtained low sensitivity, demonstrate strain-insensitive sensing...
Considerable research has been reported on developing effective active control means to suppress oscillating combustion. The typical pressure oscillation can be divided into linear growth, transition and saturation stages. In this study, a sliding mode strategy, consisting of state estimate model, disturbance observers controller, is proposed the longitudinal strategy first tested with nonlinear 0D space model as controlled plant. Results show that combined singular spectrum analysis (SSA)...
Combustion kinetic modeling is an integral part of combustion simulation, and extensive studies have been devoted to developing both high fidelity computationally affordable models. Despite these efforts, kinetics still challenging due the demand for expert knowledge optimization against experiments, as well lack understanding associated uncertainties. Therefore, data-driven approaches that enable efficient discovery calibration models received much attention in recent years, core which...
Amidst the surge in deep learning-based password guessing models, challenges of generating high-quality passwords and reducing duplicate persist. To address these challenges, we present PagPassGPT, a model constructed on Generative Pretrained Transformer (GPT). It can perform pattern guided by incorporating structure information as background knowledge, resulting significant increase hit rate. Furthermore, propose D&C-GEN to reduce repeat rate generated passwords, which adopts concept...