A5048258163- Machine Fault Diagnosis Techniques
- Machine Learning in Bioinformatics
- Engineering Diagnostics and Reliability
- Advanced SAR Imaging Techniques
- VLSI and Analog Circuit Testing
- Radar Systems and Signal Processing
- Gear and Bearing Dynamics Analysis
- Advanced Algorithms and Applications
- Non-Invasive Vital Sign Monitoring
- Engineering and Test Systems
- Microwave Imaging and Scattering Analysis
- Integrated Circuits and Semiconductor Failure Analysis
Chinese Academy of Medical Sciences & Peking Union Medical College
2025
Shanghai University
2021-2023
University of Electronic Science and Technology of China
2023
Cuffless blood pressure (BP) estimation is critical for managing growing concerns about hypertension and cardiovascular diseases. Despite recent advancements in multimodal (ECG PPG) BP methods, which have achieved varying degrees of success, several challenges remain to be addressed. These include capturing the full spectrum BPrelevant information, redundant feature spaces, handling multigrade classification. To address these issues, we propose a Multimodal Fusion Network (MuFuBP-Net),...
Fault diagnosis of rotating machinery plays an important role in modern industrial machines. In this paper, a modified sparse Bayesian classification model (i.e., Standard_SBC) is utilized to construct the fault system machinery. The features are extracted and adopted as input SBC-based system, kernel neighborhood preserving embedding (KNPE) proposed fuse features. effectiveness based on KNPE Standard_SBC validated by utilizing two case studies: rolling bearing shaft diagnosis. Experimental...
For permanent magnet DC motors (PMDCMs), the amplitude of current signals gradually decreases after motor starts. In this work, time domain features and time-frequency-domain extracted from several successive segments make up a feature vector, which is adopted for fault diagnosis PMDCMs. Many redundant will lead to decrease in efficiency increase computation cost, so it necessary eliminate that have negative effects. This paper presents novel supervised filter selection method reducing data...
The Rollout algorithm of sub-module testability modeling is proposed on the basis for problem long running time complex radar systems. First, fault diagnosis strategy based constructed and expected test cost formula derived. Next, system submodules, performance analysis. Finally, stochastic simulation experiments are used to verify feasibility optimization in different matrix dimensions. results show that, compared with traditional global algorithm, can significantly reduce time, but there a...
For permanent magnet DC motors (PMDCMs), the amplitude of current signals gradually decreases after motor starts. Only using signal features in a single segment is not conducive to fault diagnosis for PMDCMs. In this work, multi-segment feature extraction presented improving effect Additionally, support vector machine (SVM), classification and regression tree (CART), k-nearest neighbor algorithm (k-NN) are utilized construction models. The time domain extracted from several successive...
In this paper, we propose an adaptive constant false alarm detection method based on background Discrimination (BBD-CFAR) to address the issues of degraded performance in detecting clutter edges and multi-neighbor targets. The proposed utilizes non-uniform estimation maintain a rate introduces mean ratio power interval discriminate background. We use iterative approach improve probability target backgrounds. BBD-CFAR is applied backgrounds with different powers Its ability suppress...