A5067978987- Aerodynamics and Acoustics in Jet Flows
- Turbomachinery Performance and Optimization
- Model Reduction and Neural Networks
- Probabilistic and Robust Engineering Design
- Power System Reliability and Maintenance
- Speech and dialogue systems
- Heat Transfer Mechanisms
- Cleft Lip and Palate Research
- Industrial Vision Systems and Defect Detection
- Linguistics and Language Analysis
- Infrastructure Maintenance and Monitoring
- Advanced Combustion Engine Technologies
- Spectroscopy and Chemometric Analyses
- Computer Graphics and Visualization Techniques
- Aerodynamics and Fluid Dynamics Research
- Aerospace and Aviation Technology
- Cavitation Phenomena in Pumps
- Transport Systems and Technology
- Advanced Multi-Objective Optimization Algorithms
- Advanced Image Processing Techniques
Fudan University
2021-2023
Nanjing University of Aeronautics and Astronautics
2022-2023
Institute of Linguistics
2020
Beijing Normal University
2020
Tianjin University of Science and Technology
2015
Abstract This study aims to investigate public’s understanding of intelligent speech technology and evaluate the word accuracy rate recognition in Japanese Chinese translation, using simulating corpus doctor-patient communication. It is concluded from questionnaire that currently public has a high awareness demand products. The results experiments show rates are 99.4% Mandarin, 96.1% Japanese, 94.2% translation Chinese, 82.7% respectively. In Mandarin errors mainly include missing words...
In machine vision defects detection of pharmaceutical blister packaging, the traditional algorithms always do binary image processing and edge feature extraction other operations, then locate analysis through these features. These operations are very tedious inflexible. this paper, shape template matching algorithm is applied to packaging detection. The can determine whether there a positioning region sheet or pieces missing search comparing difference gray value between region. This paper...
As a core component of an aero-engine, the aerodynamic performance nacelle is essential for overall aircraft. However, direct design three-dimensional (3D) limited by complex space consisting different cross-section profiles and irregular circumferential curves. The deep manifold learning-assisted geometric multiple dimensionality reduction method combines autoencoders (AE) with strong capabilities non-linear data class function/shape function transformation (CST). A novel developed to...
Inlet separation is one of the primary sources distortion at fan entrances. High angles attack (AOAs) are critical cause inlet separation. Optimizing shape to increase angle great engineering significance. However, calculating different shapes requires extensive computation. To address this problem, an ensemble indicator, antiseparation potential index (ASPI), proposed in paper. The ASPI based on circumferential (IDC) engine under multiple AOAs. characteristics both and level comprehensively...
In order to evaluate the vehicle reliability in process of user's use, a fatigue pseudo damage evaluation model Based on axle head acceleration is proposed. acceleration, an improved MINER cumulative studied. model, influence user condition, running speed, load capacity and pavement condition discussed. The operation under use established. can be used for on-line, short-term long-term prediction reliability.
Conventional approaches of design variable generation in nacelle cases are usually proposed from either two perspectives: (1) utilizing geometric resemblance to airfoil shape; or (2) selecting flexible parametrization so that any potential geometry can be produced. However, the relationship between and aerodynamic performance objectives is not emphasized these approaches: variables likely carrying information most needed for optimization, as case cowl drag reduction. In fact, an ideal should...