A5028907215- Computer Graphics and Visualization Techniques
- Surface Treatment and Residual Stress
- 3D Shape Modeling and Analysis
- Advanced Combustion Engine Technologies
- High-Velocity Impact and Material Behavior
- Advanced Numerical Analysis Techniques
- Erosion and Abrasive Machining
- Combustion and flame dynamics
- Image and Signal Denoising Methods
- Advanced Vision and Imaging
- Non-Invasive Vital Sign Monitoring
- Biodiesel Production and Applications
- Cardiovascular Health and Disease Prevention
- Fluid Dynamics and Turbulent Flows
- Catalytic Processes in Materials Science
- Fluid Dynamics and Heat Transfer
- Magnesium Alloys: Properties and Applications
- Advanced Data Storage Technologies
- Petri Nets in System Modeling
- Geophysical and Geoelectrical Methods
- Metal and Thin Film Mechanics
- Laser-Ablation Synthesis of Nanoparticles
- Microstructure and mechanical properties
- Computational Fluid Dynamics and Aerodynamics
- ECG Monitoring and Analysis
Tencent (China)
2019-2024
Nanjing University of Information Science and Technology
2024
Shanghai Jiao Tong University
2021
China University of Petroleum, Beijing
2017-2019
University of Macau
2017-2019
Jiangsu University
2006-2018
Jilin Medical University
2017
Jilin University
2017
University of Science and Technology Beijing
2014
The pure nanocrystalline Al layer and Fe containing were fabricated on the surface of 2024 alloy by mechanical attrition treatment (SMAT) with two different mediums, ceramic balls steel balls, respectively. corrosion behaviour before after SMAT was investigated potentiodynamic polarisation curves electrochemical impedance spectroscopy (EIS). tests demonstrated that improved resistance alloy, while induced led to diminution in resistance.
Creating visualizations of multiple volumetric density fields is demanding in virtual reality (VR) applications, which often include divergent distributions mixed with geometric models and physics-based simulations. Real-time rendering such complex environments poses significant challenges for quality performance. This paper presents a novel scheme efficient real-time varying translucent global illumination (GI) effects on high-resolution binocular VR displays. Our proposes creative...
Abstract We introduce an efficient technique for recovering the vector potential in wavelet space to simulate pointwise incompressible fluids. This ensures that fluid velocities remain divergence‐free at any point within domain and preserves local volume during simulation. Divergence‐free wavelets are utilized calculate coefficients of potential, resulting a smooth with enhanced accuracy, even when input exhibit some degree divergence. accuracy eliminates need additional computational time...
<title>Abstract</title> Aiming at the problem of marine garbage types detection on YOLOv7-Tiny network, a 2-IDetect network named MGYOLOv7-Tiny is proposed to relieve contradiction between parameter quantity and accuracy. Firstly, number heads reduced balance demands accuracy improvement slimming. Secondly, local feature information exchanged by light-weight convolution GSConv. Model training-time inference-time illating with different Re-parameterization Visual Geometry Group (RepVGG)...
Summary Magnetotelluric (MT) has been widely used in oil & gas exploration and the investigation of deep earth structures. MT data interpretation generally assumes an isotropic model. However, this is sometimes inadequate, because electrical anisotropy frequently found models can lead to misinterpretations. In such cases, one needs use anisotropic model for interpretation. paper, we study forward modelling 3D with arbitrary based on a goal-oriented adaptive finite-element method. This...
Abstract We introduce a new biorthogonal wavelet approach to creating water‐tight surface defined by an implicit function, from finite set of oriented points. Our aims at addressing problems with previous methods which are not resilient missing or nonuniformly sampled data. To address the problems, our has two key elements. First, applying three‐dimensional partial integration, we derive integral formula compute coefficients without requiring function be indicator function. It can shown that...
Recently, the interest in controller synthesis for complex tasks is rapidly growing [1], [2], and most cases, environments are unknown, which limits applications of traditional control methods. In this paper, we use reinforcement learning method to learn optimally achieve under unknown environments. Specifically, model uncertain using Markov Decision Processes (MDPs). The high-level objective described by syntactically co-safe Linear Temporal Logics (scLTLs). Under such settings, propose a...
Abstract Physically based deformation simulation has been studied for many years in computer graphics. In order to simulate more complex geometric models and better meet the designer's requirements, anisotropic approaches have proposed recent years. However, most of focus on simulating orthotropic models. comparison with models, nonorthotropic ones allow objects behaviors along nonorthogonal directions. this paper, we introduce an improved approach under large deformation. The improvements...
Abstract Physically plausible fracture animation is a challenging topic in computer graphics. Most of the existing approaches focus on isotropic materials. We proposed frame‐field method for design anisotropic brittle patterns. In this case, material anisotropy determined by two parts: elastic deformation and damage mechanics. For deformation, we reformulate constitutive model hyperelastic materials to achieve adding additional energy density functions particular directions. evolution,...
Abstract Gradient‐domain compositing has been widely used to create a seamless composite with gradient close field generated from one or more registered images. The key this problem is solve Poisson equation, whose unknown variables can reach the size of if no region interest drawn explicitly, thus making both time and memory cost expensive in processing multi‐megapixel In paper, we propose an approximate projection method based on biorthogonal Multiresolution Analyses (MRA) equation. Unlike...
Large-eddy simulation (LES) of diesel spray and combustion was performed to study its improvement in the engine in-cylinder dynamics compared Reynolds-averaged simulation. For LES, dynamic structure approach used model sub-grid turbulence interaction with moving droplets spray. A multicomponent vaporization (MCV) based on continuous thermodynamics a gamma distribution describe numerous fuel components, simulate droplets. The MCV imbedded into LES framework KIVA-4 program. Using this model,...
Diesel spray and combustion in a constant-volume engine cylinder was simulated by large eddy simulation (LES) approach coupling with multicomponent vapourisation (MCV) modelling. The focused on the inclusion of interaction between fuel gas-phase turbulence flow at sub-grid scale. LES based dynamic structure model, an additional source term added to filtered momentum equation account for effect drop motion turbulence. modelling continuous thermodynamics using gamma distribution describe...
Abstract In this paper, a novel method with good numerical stability is proposed from the perspective of energy preserving to alleviate dissipations in advection step Eulerian fluid simulation. The main idea measure vorticity loss during advection, calculate lost angular kinetic scheme, and then synthesize high‐frequency incompressible details field compensate way that consistent Kolmogorov's theory, which prevents synthetic interfering existing flow. works independently scheme can be easily...
Physically based deformation simulation has been studied for many years in computer graphics. In order to simulate more complex materials and better meet the designer's requirements, anisotropic approach was proposed recent years. However, most of existing approaches focus on orthotropic materials. this paper, a general non-orthogonal constitutive model is presented behavior 3D soft objects. The exhibits different behaviors directions by constructing coordinate system with covariant...
This paper presents a novel solution for approximations to some large convolution kernels by leveraging weighted box-filtered image pyramid set. Convolution filters are widely used, but still compute-intensive real-time rendering when the kernel size is large. Our algorithm approximates kernels, such as Gaussian and cosine filters, two phases of down up sampling on GPU. The computational complexity only depends input resolution independent size. Therefore, our method can be applied...
Abstract In this paper, we propose a novel second‐order explicit midpoint method to address the issue of energy loss and vorticity dissipation in Eulerian fluid simulation. The basic idea is explicitly compute pressure gradient at middle time each step apply it velocity field after advection. Theoretically, our solver can achieve higher accuracy than first‐order solvers similar computational cost. On other hand, twice even faster implicit cost small accuracy. We have carried out large number...