A5075379587- Seismology and Earthquake Studies
- earthquake and tectonic studies
- Advanced Combustion Engine Technologies
- Advanced Aircraft Design and Technologies
- GNSS positioning and interference
- Fault Detection and Control Systems
- Anomaly Detection Techniques and Applications
- Earthquake Detection and Analysis
- Rocket and propulsion systems research
- High voltage insulation and dielectric phenomena
- Microgrid Control and Optimization
- Turbomachinery Performance and Optimization
- Distributed Control Multi-Agent Systems
- Adaptive Control of Nonlinear Systems
- Mathematical and Theoretical Epidemiology and Ecology Models
- Time Series Analysis and Forecasting
- Remote Sensing and LiDAR Applications
- Stability and Control of Uncertain Systems
- Vehicle emissions and performance
- Refrigeration and Air Conditioning Technologies
- COVID-19 epidemiological studies
- Power System Optimization and Stability
- Neural Networks Stability and Synchronization
- HVDC Systems and Fault Protection
- Power Transformer Diagnostics and Insulation
Beihang University
2019-2025
Beijing Information Science & Technology University
2025
National Marine Environmental Forecasting Center
2024
Ministry of Natural Resources
2024
Wuhan University
2024
Shenyang University of Technology
2021-2022
Naval University of Engineering
2020-2021
Hohai University
2019
Shanghai Jinyuan Senior High School
2017
Chalmers University of Technology
2014-2017
Limited by the poor transient response performance of turbochargers, dynamic aviation piston engines tends to deteriorate. In a bid enhance turbocharger's acceleration capabilities, this study scrutinizes various factors impacting its performance. Based on operational principles and process turbocharger, three types inertia—namely, aerodynamic inertia (ADI), thermal (TI), mechanical (MI) — are identified addressed for design. To begin, paper pioneers innovative definition method evaluating...
Accurate prediction of wind power generation is great significance for the efficient operation farms. However, traditional deep learning-based methods predict without simultaneously considering temporal features and spatial between variables, which leads to low accuracy. This article proposes a novel forecasting approach based on graph convolution network (GCN) multiresolution neural (CNN), combining features. In this approach, GCN merged with maximum information coefficient (MIC) proposed...
The telemetry data obtained from an on-orbit spacecraft contain important information to indicate anomaly of the spacecraft. However, large number monitoring variables and amount points, as well lack prior knowledge about due complicated structure its working conditions, pose great challenge detection. This article proposes detection algorithm based on a spatial–temporal generative adversarial network (GAN) for in data. establishes GAN-based model combining convolutional neural (CNN) long...
Voltage fault diagnosis is critical for detecting and identifying the Lithium-ion battery failure. This paper proposes a voltage algorithm based on an Equivalent Circuit Model-Informed Neural Network (ECMINN) method batteries, which aims to learn observer by embedding Model (ECM) into neural network structures. directly embeds deterministic mechanism part in ECM designs uncertain networks, takes advantage of high precision physical model strong nonlinear processing ability improve effect. In...
The highly accurate data of topography and bathymetry are fundamental to ecological studies policy decisions for coastal zones. Currently, the automatic extraction classification signal photons in zones is a challenging problem, especially surface type without auxiliary data. lack information limits large-scale bathymetric applications ICESat-2 (Ice, Cloud, Land Elevation Satellite-2). In this study, we propose photon extraction–classification method process geolocated areas from ATL03...
Detecting anomalies for multivariate time series is of great importance in modern industrial applications. However, due to the complex temporal dynamics systems, finding a distinguishable judge criterion hard, which makes accurate anomaly detection still challenging task. In order better capture anomalous features and design more informative criterion, this article presents an unsupervised generative adversarial network (GAN) detection, highlights novel active distortion transformer (ADT)...
Efficient anomaly detection in telemetry time series is of great importance to ensure the safety and reliability spacecraft. However, traditional methods are complicated train, have a limited ability maintain details, do not consider temporal-spatial patterns. These problems make it still challenge effectively identify anomalies for multivariate series. In this paper, we propose Denoising Diffusion Time Series Anomaly Detection (DDTAD), an unsupervised reconstruction-based method using...
Remaining useful life (RUL) prediction of rolling bearings brings benefits for maintenance spacecrafts. Vibration signals are widely used RUL prediction. However, under some situations such as high-speed rotation bearings, vibration quite easily disturbed by noise and might be tough to collect due inappropriate installation accelerometers. Therefore, in this paper, stator current considered health indicator bearing Based on signals, feature extraction trajectory tracking suffer two...
An effective health assessment guarantees the high accuracy of remaining useful life (RUL) prediction machinery components. The key to is indicator components, which are generally constructed by feature fusion from time and frequency domains in sensorial data. However, existing construction methods constrained due low-sampling rate susceptibility environmental disturbances To draw these issues, this paper mainly proposes a method combining condensed image coding metrics-constrained deep...
Accurate observation of nearshore waves is crucial for coastal safety. In this study, the feasibility extracting wave information from video images captured by shore-based cameras using deep learning methods was explored, focusing on inverting significant height (SWH) instantaneous images. The accuracy models in classifying wind and swell investigated, providing reliable classification results SWH inversion research. A network named ResNet-SW types with improved ResNet proposed. On basis,...
The telemetry data of spacecraft is an ultra-high dimensional time series used to indicate on-orbit operation status, and anomaly detection can effectively ensure safety reliability. Aiming at the characteristics complex correlation high data, this paper proposes a novel method based on Generative Adversarial Networks (GAN) for detection. Instead treating each variable independently, our proposed captures latent representation amongst multi-dimensional series. For normal GAN-based obtain...
Abstract Airborne LiDAR bathymetry offers low cost and high mobility, making it an ideal option for shallow-water measurements. However, due to differences in the measurement environment laser emission channel, received waveform is difficult extract using a single algorithm. The choice of suitable processing method thus extreme importance guarantee accuracy bathymetric retrieval. In this study, we use wavelet-denoising denoise subsequently test four algorithms denoised-waveform processing,...
Abstract Airborne LiDAR bathymetry offers low cost and high mobility, making it an ideal option for shallow-water measurements. However, due to differences in the measurement environment laser emission channel, received waveform is difficult extract using a single algorithm. The choice of suitable processing method thus extremely important guarantee accuracy bathymetric retrieval. In this work, we use wavelet-denoising denoise then test four algorithms denoised-waveform processing:...