A5100603465- Fluid Dynamics and Turbulent Flows
- Computational Fluid Dynamics and Aerodynamics
- Probabilistic and Robust Engineering Design
- Wind and Air Flow Studies
- Model Reduction and Neural Networks
- Plasma and Flow Control in Aerodynamics
- Video Surveillance and Tracking Methods
- Nuclear Engineering Thermal-Hydraulics
- Rough Sets and Fuzzy Logic
- Advanced Computational Techniques and Applications
- Nonlinear Photonic Systems
- Advanced Fiber Laser Technologies
- Gas Dynamics and Kinetic Theory
- Adaptive Control of Nonlinear Systems
- Human Pose and Action Recognition
- Anomaly Detection Techniques and Applications
- Lattice Boltzmann Simulation Studies
- Remote Sensing and Land Use
- Advanced Control Systems Optimization
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Multi-Criteria Decision Making
- Fault Detection and Control Systems
- Quantum Chromodynamics and Particle Interactions
- Industrial Technology and Control Systems
Air Force Engineering University
2023-2024
University of Jinan
2009-2024
CCCC Highway Consultants (China)
2011-2024
Beihang University
2021-2023
Xi'an University of Technology
2023
Wuhan University
2004-2020
China Electronics Technology Group Corporation
2020
Shandong University
2018-2019
Xinjiang Uygur Autonomous Region Disease Prevention and Control Center
2017
Beijing Union University
2009-2012
In this paper, a continuous Q-learning algorithm is deployed to optimize the control strategy of trailing-edge airfoil flow separation at chord-based Reynolds number 2×105. With plasma synthetic jets issued middle chord and hot wire placed in separated shear layer acting as actuator sensor, respectively, high-speed reinforcement learning an interaction frequency 500 Hz realized by field-programmable gate array. The results show that control, controller only needs several seconds elevate...
Due to the strong stress resistance, causing food spoilage and toxin production of spore‐forming Bacillus species, their presence will be detrimental quality assurance dairy products. Impacts five strains ( licheniformis H1, B. H2, subtilis Z2, cereus Z6‐1 Z6‐2) originated from products on pasteurised yoghurt were investigated. Physiological characteristics strains, including growth, acid production, proteolysis, amylase extracellular polysaccharide biofilm formation, analysed. Additionally,...
Turbulence modeling based on the Reynolds-averaged Navier–Stokes (RANS) method has been widely applied in industry, but its performance some complex flows is far from satisfactory. The improvement of turbulence models traditional framework not made breakthrough progress for decades. In this study, a data-driven reconstruction Reynolds stress discrepancies used to aid models, with stresses shear-stress transport model being modified eigenspace. large eddy simulation (LES) dataset set bump...
The Reynolds Averaged Navier-Stokes (RANS) models are still the workhorse in current engineering applications due to its high efficiency and robustness. However, closure coefficients of RANS turbulence determined by model builders according some simple fundamental flows, suggested values may not be applicable complex especially supersonic jet interaction flow. In this work, Bayesian method is employed recalibrate Spalart-Allmaras (SA) improve performance problem quantify uncertainty wall...
In this experimental study, a value-based reinforcement learning algorithm (deep Q-network, DQN) is used to control the flow separation behind backward facing step at Reynolds number of 2.9 × 104. The forced by dielectric barrier discharge (DBD) plasma actuator pasted upstream edge, and feedback information zone provided hotwire sensor submerged in downstream shear layer. law represented deep neural network implemented on field programable gate array (FPGA), able execute real-time frequency...
Cavity oscillation is a significant and challenging issue in aerospace engineering. Surface dielectric barrier discharge (SDBD) plasma actuators can induce electrical body forces instantaneously the flow, have been proven to be effective suppressing noise caused by cavity oscillation. This research focuses on steady unsteady operation modes of SDBD actuators. A systematic parametric study flow conducted experimentally, characteristics are analyzed through acoustic particle image velocimetry...
Although active flow control based on deep reinforcement learning (DRL) has been demonstrated extensively in numerical environments, practical implementation of real-time DRL experiments remains challenging, largely because the critical time requirement imposed data acquisition and neural-network computation. In this study, a high-speed field-programmable gate array (FPGA) -based experimental (FeDRL) framework is developed, capable achieving frequency 1–10 kHz, two orders higher than that...
Abstract Taking body forces into account is not new for the lattice Boltzmann method, yet most of existing approaches can only treat steady and uniform forces. To manage situations with time‐ space‐dependent or source terms, this paper proposes a approach through theoretical derivation numerical verification. The method by attaching an extra term to equation still used, but expression modified. It modified that achieves particularity approach. This introduce unsteady non‐uniform momentum...
In the inlet passage of a hypersonic vehicle, multi-channel shock waves inevitably interact with boundary layer, producing complex wave/boundary layer interactions (SWBLIs). The flow separation caused by these significantly decreases intake efficiency and may prevent from starting. typical interaction mode is through double incident SWBLIs. Therefore, it necessary to study characteristics SWBLIs identify relevant control techniques. this paper, are first examined, then results an...
The Bayesian uncertainty quantification technique has become well-established in turbulence modeling over the past few years. However, it is computationally expensive to construct a globally accurate surrogate model for inference high-dimensional design space, which limits complex flow configurations. Borrowing ideas from stratified sampling and inherited sampling, an adaptive refinement approach proposed this work, concentrates on asymptotically improving local accuracy of...
The Shock Wave-Boundary Layer Interaction (SWBLI) flow generated by compression corner widely occurs in engineering. As one of the primary methods engineering, Reynolds Averaged Navier-Stokes (RANS) usually cannot correctly predict strong SWBLI flows. In addition to defects eddy viscosity assumption, uncertainty closure coefficients RANS models often significantly impacts simulation results. This study performs parametric sensitivity analysis and Bayesian calibration on Menter k-ω...
Brucellosis is a worldwide zoonotic disease caused by Brucella spp. In China, brucellosis recognized as reemerging mainly melitensis specie. To better understand the currently endemic B. strains in three genotyping methods were applied to 110 obtained past several years. By MLVA genotyping, five MLVA-8 genotypes identified, among which 42 (1-5-3-13-2-2-3-2) was predominant genotype, while genotype 63 (1-5-3-13-2-3-3-2) and novel of 1-5-3-13-2-4-3-2 second frequently observed. MLVA-16...
Accurate predictions of flow separation are important for aerospace design, flight accident avoidance, and the development fluid mechanics. However, complexity process makes accurate challenging all known Reynolds-averaged Navier–Stokes (RANS) methods, underlying mechanism action remains unclear. This paper analyzes specific reasons defective turbulence models applied to separated flows, explores physical properties that impact predictions, investigates their mechanisms. Taking Menter SST...
Shock wave-boundary layer interactions (SWBLIs) have a tremendous influence on the performance of hypersonic vehicles. For numerical simulation such engineering flows, Reynolds averaged Navier-Stokes (RANS) still occupies an irreplaceable role. However, parameters turbulence models in RANS substantial uncertainties, which impact reliability results. Thus, aim present study is to conduct uncertainty analysis shear-stress transport (SST) model for SWBLIs. In current work, quantification was...
Aerodynamic/stealth design is becoming an important factor in the advanced airfoil design. In this work, a supervised machine learning method proposed for aerodynamic and stealth integrated The conditional generative adversarial network (CGAN) constructed multidisciplinary of airfoil. Then, generator discriminator simply using deep neural have good robustness stability training. CGAN model also shows generalization capability test set, with less than 1% error fitting to profile data,...
Shock-wave/boundary-layer interaction (SWBLI) is a fundamental scientific problem that restricts the breakthrough of high-speed flight technology. Owing to complex shock structure and boundary-layer separation in region, widely used eddy-viscosity turbulence models have large uncertainties model errors simulation such flows. To make more reasonable aerodynamic/aerothermal predictions identify critical parameters affecting prediction results, Bayesian uncertainty quantification sensitivity...