- Turbomachinery Performance and Optimization
- Advanced Multi-Objective Optimization Algorithms
- Photoacoustic and Ultrasonic Imaging
- Thermography and Photoacoustic Techniques
- Heat Transfer Mechanisms
- Nanoplatforms for cancer theranostics
- Probabilistic and Robust Engineering Design
- Optimal Experimental Design Methods
- Combustion and flame dynamics
- Fluid Dynamics and Turbulent Flows
- Heat Transfer and Optimization
- Optical Imaging and Spectroscopy Techniques
- Advanced Aircraft Design and Technologies
- Image Enhancement Techniques
- Metaheuristic Optimization Algorithms Research
- Gaussian Processes and Bayesian Inference
- COVID-19 diagnosis using AI
- Cavitation Phenomena in Pumps
- Optical Coherence Tomography Applications
- Cyclone Separators and Fluid Dynamics
- Photodynamic Therapy Research Studies
- Model Reduction and Neural Networks
- Aerodynamics and Fluid Dynamics Research
- Heat transfer and supercritical fluids
- 3D Shape Modeling and Analysis
Xi'an Jiaotong University
2014-2024
Tianjin University
2023-2024
Hubei University of Science and Technology
2024
Shanghai Jiao Tong University
2017-2023
Guangdong Ocean University
2023
Nanyang Technological University
2019-2022
Tianjin University of Science and Technology
2021-2022
Acoustic-resolution photoacoustic microscopy (ARPAM) plays an important role in studying the microcirculation system of biological tissues with deep penetration. High lateral resolution ARPAM is achieved by using a high numerical aperture acoustic transducer. The deteriorated out-of-focus region can be alleviated synthetic focusing technique (SAFT). Previously, we reported three-dimensional (3D) deconvolution to improve both and axial resolutions focus region. In this study, present our...
The point-by-point scanning mechanism of photoacoustic microscopy (PAM) results in low-speed imaging, limiting the application PAM. In this work, we propose a method to improve quality sparse PAM images using convolutional neural networks (CNNs), thereby speeding up image acquisition while maintaining good quality. CNN model utilizes attention modules, residual blocks, and perceptual losses reconstruct image, which is mapping from 1/4 or 1/16 low-sampling latent fully-sampled one. trained...
The turbomachinery cascades design is a typical high dimensional computationally expensive and black box problem, thus metamodel-based optimization data mining method proposed programed in this work, which intended to gain knowledge of space except for optimal solutions. combines Kriging-based global algorithm with techniques self-organizing map (SOM), analysis variance (ANOVA), parallel axis. NACA Rotor 37, axial transonic rotor blade, selected the research. Through SOM analysis, overall...
A miniature all-optical probe for high-resolution photoacoustic (PA)-ultrasound (US) imaging using a large synthetic aperture is developed. The consists of three optical fibers PA excitation, US generation, and detection acoustic waves, respectively. fiber excitation has numerical (NA) wide-angle laser illumination. On the other hand, with carbon black-polydimethylsiloxane composite coated on end face used transmission through laser-US conversion. Both excited backscattered signals are...
We present a miniature probe capable of both optical-resolution (OR) and acoustic-resolution (AR) photoacoustic microscopy. A gradient-index-lens fiber multimode are used to deliver light for OR AR illumination, respectively. The achieves lateral resolution 3.1 μm mode 46-249 (at depth 1.2-4.3 mm) mode, size the attains 3.7 mm in diameter, which can be endoscopic applications. In vivo imaging several different parts mouse demonstrates excellent ability probe.
Photoacoustic endoscopy (PAE) is a promising tool for the detection of atherosclerotic plaque. In this work, we propose novel design side-viewing PAE probe based on synthetic aperture focusing technique (SAFT) to enable high transverse resolution over large depth focus (DOF) along radial direction. A point-like ultrasonic detector used ensure wide angle and thus SAFT. We first perform numerical simulation optimize design, which involves placement diameter reflection rod mirror. Then,...
An automated three-dimensional multi-objective optimization and data mining method is presented by integrating a self-adaptive differential evolution algorithm (SMODE), 3D parameterization for blade profile meridional channel, Reynolds-averaged Navier–Stokes (RANS) solver technique of self-organizing map (SOM). Using this method, redesign high pressure ratio centrifugal impeller conducted. After optimization, 16 optimal Pareto solutions are obtained. Detailed aerodynamic analysis indicates...
We present a miniature fiber-optic ultrasound transmitter for generating high-intensity focused (HIFU) based on photoacoustic transduction. The HIFU device consists of fiber and lens. develop simple fabrication procedure making the lens, which is coated with candle soot nanoparticles-polydimethylsiloxane composites. used to deliver pulsed laser excitation, facilitates use by eliminating need free-space optical alignment. (6.5 mm in diameter) produces acoustic pressures up >30 MPa peak...
Abstract The expensive computational cost is always a major concern for reliability-based design optimization (RBDO) of complex problems. performance RBDO can be lowered by the inaccuracy reliability analysis (RA) which caused multiple local optimums and points in highly non-linear space. In order to reduce burden guarantee accuracy RA (and thus improve performance), global algorithm adopting an improved constraint boundary sampling (GRBDO-ICBS) method proposed. Specifically, GRBDO-ICBS...
Abstract To solve the turbine design optimization problems efficiently, surrogate-based algorithms are frequently used. further reduce cost of design, multi-fidelity surrogate (MFS)-based is proposed by researchers, who resort to augmenting small number expensive high-fidelity (HF) samples a large portion low-fidelity (LF) but cheap in modeling and process. Nonetheless, according our observations, MFS-based sometimes can only have better convergence rate at early stage process, yielding...
Photoacoustic endoscopy (PAE), an ideal approach to enable internal organ and intravascular imaging, has adopt a tailored scanner or scanning mechanism for 2-D 3-D imaging. Forward-view found medical applications in gastrointestinal cystoscopy. Although recent works have demonstrated rotational by using micromotor joint, they are not applicable forward-view PAE. A micro-electro-mechanical systems mirror been used PAE but suffers from the large probe diameter. Optical fiber bundles either...
Abstract With the ever-increasing aerodynamic and thermal loads, endwalls of modern gas turbines have become critical areas that are susceptible to manufacturing operational uncertainties, making them highly prone failures. Therefore, it is vital importance quantify impacts input uncertainties on aero-thermal performance endwalls. First, based Kriging surrogate, an efficient uncertainty quantification (UQ) method suitable for expensive computational fluid dynamics (CFD) problems proposed. By...
Photoacoustic microscopy (PAM) is a noninvasive imaging technique and excellent to study structural functional changes in the microcirculation. In this work, lipopolysaccharide (LPS)-induced inflammation model mice noninvasively evaluated by PAM. PAM used image microvascular for 8 hours after LPS with different concentrations applied. Quantitative analysis of five vessel parameters conducted, which shows that rate reduction microvasculature highly dependent on applied concentrations. For...
A novel all-optical double gradient-index (GRIN) lens optical-resolution photoacoustic microscopy (OR-PAM), termed as DGL-PAM, is demonstrated. The miniature probe consists of a single-mode fiber and GRIN lenses for optical focusing Fabry-Perot sensor ultrasound detection. new design simple realizes high resolution with long working distance (WD) by virtue the lenses. overall size 2.7 mm in diameter. High lateral 3.7 μm (at 532 nm laser wavelength) WD 5.5 are achieved. In vivo OR-PAM mouse...
Many real-world problems, such as airfoil design, involve optimizing a black-box expensive objective function over complex structured input space (e.g., discrete or non-Euclidean space). By mapping the into latent of dozens variables, two-stage procedure labeled generative model based optimization (GMO) in this paper, shows promise solving problems. However, dimension GMO is hard to determine, which may trigger conflicting issue between desirable solution accuracy and convergence rate. To...
The optimization and knowledge discovery of a low-aspect-ratio cascade was conducted. techniques nonaxisymmetric endwall contouring, section profiling, compound lean were combined for parameterization. Their effects on performance investigated through analysis flow details well-designed scatter plot. It is found that contouring helpful to reduce the profile loss while also inhibiting secondary flow. Section profiling an effective technique loss. And structure responsible multimodality...