A5102808040- Smart Grid and Power Systems
- Power Systems and Technologies
- Vehicle Routing Optimization Methods
- High voltage insulation and dielectric phenomena
- Air Traffic Management and Optimization
- Advanced Neural Network Applications
- Aviation Industry Analysis and Trends
- Advanced Measurement and Detection Methods
- Scheduling and Timetabling Solutions
- High-Voltage Power Transmission Systems
- Advanced Computational Techniques and Applications
- Power Line Communications and Noise
- Video Surveillance and Tracking Methods
- Advanced Sensor and Control Systems
- Recommender Systems and Techniques
- Industrial Vision Systems and Defect Detection
- Power Systems Fault Detection
- Infrastructure Maintenance and Monitoring
- Advanced Technology in Applications
- Image Retrieval and Classification Techniques
- Infrared Target Detection Methodologies
- Smart Grid Security and Resilience
- Advanced Manufacturing and Logistics Optimization
- Image Enhancement Techniques
- Vehicle License Plate Recognition
Zhongshan Hospital
2022-2025
Fudan University
2022-2025
Shenyang University
2024
Nanjing University of Aeronautics and Astronautics
2009-2024
Hangzhou Dianzi University
2024
Shanghai Electric (China)
2016-2023
Southeast University
2023
Nanjing University of Science and Technology
2021-2022
Northwest A&F University
2022
China Southern Power Grid (China)
2020-2022
Introduction: Laennec’s capsule serves as a critical anatomical landmark in liver resection. Despite its potential, lack of large-scale prospective studies limits the widespread use Laennec approach for minimally invasive hepatectomy. This multicenter cohort study aimed to compare outcomes traditional and approaches hepatectomy across multiple centers China. Methods: A total 445 patients from 11 were included, with 339 undergoing 106 receiving approach. Intraoperative parameters such...
Abstract Ship detection is a crucial task for waterway surveillance and channel optimization, especially in close proximity to the shore. However, detecting ship visible image‐based remains challenge due limited nature of image datasets. To address this issue, Inland Ships Data Set (ISDS) constructed facilitate research on identification. On other hand, most methods struggle accurately identify ships that are small size. Therefore, model proposed employs multi‐scale weighted feature fusion...
In orchard environments, negative obstacles such as ditches and potholes pose significant safety risks to robots working within them. This paper proposes a obstacle detection method based on LiDAR tilt mounting. With the tilted at 40°, blind spot is reduced from 3 m 0.21 m, ground point cloud density increased by an order of magnitude. Based geometric features laser clouds (such rear wall height density, spacing jump between points), for detecting presented. establishes mathematical model...
An intelligent charging load forecasting method of the electric vehicle plays a vital role in power generation and market optimization. In this paper, vehicles have been categorized as buses, taxis, private cars buses. The probabilistic models influencing factors are investigated, then calculation for different types attained. According to predicted results ownership, is calculated by use Monte Carlo Simulation extraction both initial state time. Finally, deep learning algorithm LSTM (Long...
Unmanned aerial vehicles (UAVs) are important in modern war, and object detection performance influences the development of related intelligent drone application. At present, target categories UAV tasks diversified. However, lack training samples novel will have a bad impact on task. same time, many state-of-the-arts not suitable for images due to particularity perspective large number small targets. In this paper, we design fast few-shot detector It adopts idea anchor-free fully...
With the fast development of smart grid technology in china, more and attentions were paid to condition-based maintenance high voltage equipment. To realize objectives, standardizations online monitoring system hierarchy structure function are key steps pilot substation project. According relevant published standards, technical guides solutions, this paper first introduces basic substation. After that, detail terms about device, comprehensive unit, unit communication protocols covered...
Traffic flow prediction is a significant component for the new generation intelligent transportation. In field of air transportation, accurate airport flight can help airlines schedule flights and provide decision-making basis resource allocation. With Deep Learning technology, this paper focuses on characteristics easily disturbed by environmental factors, studies spatiotemporal dependence between flows, predicts distribution flows from network level. We proposed deep learning architecture...
Abstract High-quality power equipment is the basis for ensuring safe operation of grid and improving reliability supply. In actual operation, although some electrical can continue to be used, abnormal or hidden dangers will affect safety people, grid, reliable economic equipment, output life span, quality. Therefore, research on multi-dimensional analysis same batch substation identification suspected family defects frequent great significance company’s management defect management. The...
Semantic segmentation based on deep learning has very broad application scenarios in the field of computer vision. DeepLab v3+ is currently a widely used semantic framework. This paper uses model TensorFlow to semantically segment foreign objects airport runway. Experimental results show that can accurately classify at pixel level, which effectively improves accuracy object detection and reduces risk accidental aircraft take-off landing.
Wind power has gained wide popularity due to the increasingly serious energy and environmental crisis. However, severe operational conditions often bring faults failures in wind turbines, which may significantly degrade security reliability of large-scale farms. In practice, accurate efficient fault detection diagnosis are crucial for safe reliable system operation. This work develops an effective deep learning solution using a convolutional neural network address said problem. addition,...
Under the background of Energy Internet, ever-growing scale electric power system has brought new challenges and opportunities. Numerous categories measurement data, as cornerstone communication, play a crucial role in security stability system. However, present sampling transmission equipment inevitably suffers from data missing, which seriously degrades stable operation state estimation. Therefore, this paper, we consider load an example first develop missing detection algorithm terms...
The SSD algorithm is currently a widely used target detection algorithm. In view of the small size common foreign objects on airport runways and low accuracy processing objects, this paper applies feature fusion method in model to construct new object model. detected self-built runway body data set, result mAP 5% higher than original network.
OPPC (Optical Phase Conductor) is a particular type of electric optical cables which composite the fiber unit into structure traditional phase lines. The special design fully leverages power system's own line resources and achieves dual functions transmission communication simultaneously, particularly in distribution networks. Furthermore, Raman time domain reflectometry (ROTDR) based distributed temperature sensing (DTS) system integrates with OPPC, that to plant single or several multimode...
Abstract Facial recognition technology and related research have matured over time, but in the field of individual animal is still very limited. Therefore, this article focuses on identification cashmere goats with similar characteristics. First, single shot multibox detector network was used to process dataset. Next, transfer learning applied learn characteristics goats, as well loss function composed Triplet Loss Label Smoothing CrossEntropy function. The result fused by multiple different...
This paper discussed the application of online monitoring system based on IEC61850 standard in smart substation. Firstly, functions and performance requirements have been discussed, functional model interface established; Secondly, schema has designed for Furthermore, according to practical pilot project, validity models verified, a summary given.
Data of power transmission and transformation in smart grid demonstrates the characteristics big data. First all, article listed types data, introduced source data explained how to gain data; then analyzed aspects volume, velocity, variety value which were well-known field
At present, the distribution network access lacks uniform standard, which makes communication protocols of manufacturers are not compatible with each other, and dues to different services scenarios need adapt devices systems. These reasons lead extremely difficult manage maintain system, seriously hampered process comprehensive information power distribution. In this paper, a wireless device for adaptive multi interface protocol is designed in allusion problem existing technology. This has...
In this paper, we present a neural network (InteractionNN) for sparse predictive analysis where hidden features of data can be learned by multilevel feature interaction. To characterize interaction features, InteractionNN consists three modules, namely, nonlinear pooling, layer-lossing, and embedding. Nonlinear pooling (NI pooling) is hierarchical structure and, shortcut connection, constructs low-level interactions from basic dense to elementary features. Layer-lossing feed-forward...