Search

Your search keyword '"Xu, Chengji"' showing total 36 results
Start Over Author "Xu, Chengji"
36 results on '"Xu, Chengji"'

21. Wake2Wake: Feature-Guided Self-Supervised Wave Suppression Method for SAR Ship Wake Detection

Author

Xu, Chengji, Wang, Qingsong, Wang, Xiaoqing, Chao, Xiaopeng, and Pan, Bo

Abstract

Sea clutter and inherent speckle noise in synthetic aperture radar (SAR) images can pose challenges to accurate sea surface target detection, especially for phenomena such as ship wakes reliant on efficient feature extraction. Traditional denoising methods require manual tradeoffs between denoising effects and detail retention. Supervised denoising methods based on deep learning demand a substantial number of real noisy-clean image pairs for training, coupled with specific parameter settings and labeled data amounts. In response to these challenges, this article introduces Wake2Wake, a self-supervised denoising method aimed at enhancing the performance of existing deep learning-based ship wake detectors. The method incorporates a novel ship wake awareness (SWA) block designated to address the distinctive features of turbulent and Kelvin wakes. Furthermore, to overcome the source imbalance problem in the dataset, simulated wake data are integrated into the training process. This not only mitigates dataset imbalances but also significantly improves both denoising and detection performance. The experimental results indicate that Wake2Wake improves the accuracy of Rotated RepPoints by 3.6 mAP and S2A-Net by 2.6 mAP on the OpenSARWake dataset, respectively. The proposed approach achieves varied extents of improvement, showcasing its potential in mitigating sea clutter and enhancing feature extraction, especially in detecting SAR ship wakes.

Published
2024
Full Text
View/download PDF

28. Improved power allocation strategy of hybrid energy storage system in islanding DC micro‐grid

Author

Li Yuping, Xu Chengji, Wang Yuting, Chang Xiaoyong, and Chen Fufeng

Subjects
Supercapacitor, Computer science, business.industry, Voltage control, 020208 electrical & electronic engineering, General Engineering, Electrical engineering, Micro grid, Energy Engineering and Power Technology, Hybrid energy, 02 engineering and technology, Power (physics), Computer data storage, 0202 electrical engineering, electronic engineering, information engineering, Islanding, Transient response, business, Software
Published
2017

29. Grid-Connected Photovoltaic System Based on Tapped-Inductor Quasi-Z-Source Inverter and Its Low Voltage Ride-Through Control Strategy

Author

Zhao Chengbin, Sun Bo, Xie Bangpeng, Xu Chengji, Lu Hailing, and Zhang Hua

Subjects
business.industry, Computer science, 020209 energy, 020208 electrical & electronic engineering, Photovoltaic system, Electrical engineering, 02 engineering and technology, Power factor, Booster (electric power), Voltage sag, 0202 electrical engineering, electronic engineering, information engineering, Grid-connected photovoltaic power system, Low voltage ride through, business, Low voltage, Z-source inverter
Abstract

Compared with traditional Z-source and quasi-Z-source inverter, tapped-inductor quasi-Z-source inverter (TL-qZSI) has higher booster multiple, lower capacitance stress and higher conversion efficiency. Therefore, it is very suitable for grid-connected photovoltaic power generation system. This paper analyzes the voltage boosting theory of the TL-qZSI and applies it into grid-connected photovoltaic power generation system. For the imbalanced power grid voltage sag situation, this paper puts forward a low voltage ride-through (LVRT) control strategy. System realizes the unity power factor running and has a good LVRT ability under imbalanced power grid voltage sag. At last, the simulation and experimental results are shown to verify the correctness of the control method.

Published
2018

30. Low-cycle fatigue behavior of extruded Al-7Zn-2Mg-1.5Cu-0.2Sc-0.1Zr alloy at room and low temperatures

Author

Zhu Qingyan, Chen Li-jia, Xu Chengji, Li Feng, and Che Xin

Subjects
Alloy, chemistry.chemical_element, Plasticity, engineering.material, Specific strength, Amplitude, chemistry, Dynamic loading, Aluminium, lcsh:TA1-2040, Free surface, engineering, Hardening (metallurgy), Composite material, lcsh:Engineering (General). Civil engineering (General)
Abstract

Due to the low density and high specific strength, aluminum alloys have been considered for automotive and aerospace applications. The aluminum components usually service in the conditions of low temperature and dynamic loading. Therefore, the research on the low temperature fatigue performances of Al alloys has great significance. The lowcycle fatigue tests for the extruded Al-7Zn-2Mg-1.5Cu-0.2Sc-0.1Zr alloy subjected to solution plus aging treatment have been conducted at 25°C and -40°C, respectively. The strain ratio and cyclic frequency applied in the low-cycle fatigue test were -1 and 0.5Hz, respectively. The experimental results show that at 25°C, the alloy exhibits the cyclic hardening at the total strain amplitudes of 1.0% and 1.2%, and the cyclic stabilization at the total strain amplitudes of 0.4%, 0.6% and 0.8%. At -40°C, however, the cyclic stability is observed during whole fatigue deformation at the total strain amplitudes of 0.4%, 0.5%, 0.6%, 0.7% and 0.8%. The relationship between the elastic strain amplitude, plastic strain amplitude and reversals to failure can be described by Basquin and Coffin-Manson equations, respectively. In addition, the observation results of fatigue fracture surfaces reveal that the cracks initiate at the free surface of fatigue specimen and propagate in a transgranular mode.

Published
2018

32. A New Multi-Scale Convolutional Model Based on Multiple Attention for Image Classification.

Author

Yang, Yadong, Xu, Chengji, Dong, Feng, and Wang, Xiaofeng

Subjects
MULTISCALE modeling, ARTIFICIAL neural networks, COMPUTER vision, COMPUTER systems, INTRA-aortic balloon counterpulsation, MACHINE learning
Abstract

Computer vision systems are insensitive to the scale of objects in natural scenes, so it is important to study the multi-scale representation of features. Res2Net implements hierarchical multi-scale convolution in residual blocks, but its random grouping method affects the robustness and intuitive interpretability of the network. We propose a new multi-scale convolution model based on multiple attention. It introduces the attention mechanism into the structure of a Res2-block to better guide feature expression. First, we adopt channel attention to score channels and sort them in descending order of the feature's importance (Channels-Sort). The sorted residual blocks are grouped and intra-block hierarchically convolved to form a single attention and multi-scale block (AMS-block). Then, we implement channel attention on the residual small blocks to constitute a dual attention and multi-scale block (DAMS-block). Introducing spatial attention before sorting the channels to form multi-attention multi-scale blocks(MAMS-block). A MAMS-convolutional neural network (CNN) is a series of multiple MAMS-blocks. It enables significant information to be expressed at more levels, and can also be easily grafted into different convolutional structures. Limited by hardware conditions, we only prove the validity of the proposed ideas through convolutional networks of the same magnitude. The experimental results show that the convolution model with an attention mechanism and multi-scale features is superior in image classification. [ABSTRACT FROM AUTHOR]

Published
2020
Full Text
View/download PDF

33. An improved quasi-Z-source inverter for grid-connected photovoltaic system

Author

Zhao Chengbin, Sun Bo, Xu Chengji, Li Wen, and Lu Hailing

Subjects
Engineering, business.industry, 020209 energy, 020208 electrical & electronic engineering, Photovoltaic system, Electrical engineering, Topology (electrical circuits), 02 engineering and technology, Inductor, Stress (mechanics), 0202 electrical engineering, electronic engineering, information engineering, Grid-connected photovoltaic power system, Inverter, business, Z-source inverter, Voltage
Abstract

An improved quasi-Z-source inverter for grid-connected photovoltaic system is proposed in this study. Two tapped-inductors are substituted for the inductors of quasi-Z-source inverter (qZSI) respectively. Compared with the classic topology, the proposed topology can not only arise the boost ability but also make the voltage stress be decreased. The operation principle and voltage stress of the proposed topology is analyzed theoretically and compared with that of qZSI and switched-inductor quasi-Z-source inverter (SL-qZSI). The results of simulations and experiments are shown to validate the feasibility of the proposed topology.

Published
2016

Catalog

Books, media, physical & digital resources
See catalog results