A5076579785- Algebraic structures and combinatorial models
- Advanced Topics in Algebra
- Vehicle Routing Optimization Methods
- Advanced Manufacturing and Logistics Optimization
- Optimization and Packing Problems
- Nonlinear Waves and Solitons
- Collaboration in agile enterprises
- Advanced Neural Network Applications
- HVDC Systems and Fault Protection
- Advanced Combinatorial Mathematics
- Robotics and Sensor-Based Localization
- Vehicle License Plate Recognition
- Industrial Technology and Control Systems
- Domain Adaptation and Few-Shot Learning
- Robotic Path Planning Algorithms
- Urban and Freight Transport Logistics
- Smart Agriculture and AI
- High-Voltage Power Transmission Systems
- Metaheuristic Optimization Algorithms Research
- Transportation and Mobility Innovations
- Homotopy and Cohomology in Algebraic Topology
- Power Systems Fault Detection
- Mobile Crowdsensing and Crowdsourcing
- Mobile Agent-Based Network Management
- Industrial Vision Systems and Defect Detection
South China University of Technology
2015-2025
Universidad del Noreste
2022-2025
Merck & Co., Inc., Rahway, NJ, USA (United States)
2025
Beijing Information Science & Technology University
2019-2024
China University of Mining and Technology
2023-2024
Shanghai Institute of Technical Physics
2024
Invictus Medical (United States)
2024
Zhejiang University of Technology
2024
Hangzhou City University
2024
National Cheng Kung University Hospital
2024
Using multi-source sensing data based on the Internet of Things (IoT) with artificial intelligence and big processing technology to achieve predictive maintenance mechanical equipment can remarkably improve service life machine reduce labor costs when diagnosing faults, it has become a highly relevant research topic. In this paper, fusion models algorithms are studied discussed. First, Joint Directors Laboratories (JDL) model Hierarchical compared analyzed. Then, various types Neural...
In recent years, Deep Learning (DL), such as the algorithms of Convolutional Neural Networks (CNN), Recurrent (RNN) and Generative Adversarial (GAN), has been widely studied applied in various fields including agriculture. Researchers agriculture often use software frameworks without sufficiently examining ideas mechanisms a technique. This article provides concise summary major DL algorithms, concepts, limitations, implementation, training processes, example codes, to help researchers gain...
The rapid development of Internet Things technology has promoted the popularization Vehicles, and its safety reliability have become focus intelligent transportation system research. Vehicle-road collaboration relies on collaborative computing storage resources vehicle on-board unit (OBU), which are usually limited. When in edge area needs to do tasks such as driving, but own insufficient. Therefore, it other from idle vehicles road side (RSU). This resource sharing can get additional...
Abstract Background Pneumococcal disease (PD) is a group of infections that can have significant yet poorly understood impact on health-related quality life (HRQoL). This meta-analysis aims to quantify the HRQoL PD. Methods Original research studies reporting health utility, disutility, quality-adjusted year (QALY), or QALY decrement associated with PD were identified through global targeted literature review in MEDLINE (March 2023). Health utilities and QALYs converted decrements per...
Multi-source domain adaptation (MSDA) plays an important role in industrial model generalization. Recent efforts regarding MSDA focus on enhancing multi-domain distributional alignment while omitting three issues, e.g., the class-level discrepancy quantification, unavailability of noisy pseudo labels, and source transferability discrimination, potentially resulting suboptimal adaption performance. Therefore, we address these issues by proposing a prototype aggregation method that models...
Herein, to accurately predict tool wear, we proposed a new deep learning network—that is, the IE-Bi-LSTM—based on an informer encoder and bi-directional long short-term memory. The IE-Bi-LSTM uses part of model capture connections globally extract feature sequences with rich information from multichannel sensors. In contrast methods using CNN RNN, this could achieve remote extraction parallel computation long-sequence-dependent features. adopts attention distillation layer increase...
Tool wear prediction can ensure product quality and production efficiency during manufacturing. Although traditional methods have achieved some success, they often face accuracy real-time performance limitations. The current study combines multi-channel 1D convolutional neural networks (1D-CNNs) with temporal (TCNs) to enhance the precision of tool prediction. A 1D-CNN architecture is constructed extract features from multi-source data. Additionally, a TCN utilized for time series analysis...
Conventional zero-sequence current (ZSC) protection relays for low-resistance grounded systems (LGSs) are confronting challenges due to the risk of multiple single-phase grounding faults (MSGFs) and ongoing increase in penetration inverter-interfaced distributed generators (IIDGs). Such issues concerning false or failure tripping require a reconsideration traditional solutions. This paper presents novel adaptive ZSC scheme using current-compensation method LGSs with IIDGs. The mechanism...
Because unmanned forklifts need to recognise and locate pallets in warehouses, a detection algorithm based on deep learning framework was proposed. First, the authors collected large number of pictures including people pallet real warehouse marked corresponding label build logistics database. Second, object single shot multibox detector is improved trained by In prediction phase, network combines multiscale feature maps with different resolution, which enhances adaptability task. Third, an...
Vibration sensing data is an important resource for mechanical fault prediction, which widely used in the industrial sector. Artificial neural networks (ANNs) are tools classifying vibration data. However, their basic structures and hyperparameters must be manually adjusted, results prediction accuracy easily falling into local optimum. For with high levels of uncertainty, it difficult ANN to obtain correct results. Therefore, we propose a multifeature fusion model based on Dempster-Shafer...
Abstract Nowadays, collaborative filtering recommender systems have been widely deployed in many commercial companies to make profit. Neighborhood‐based (CF) is common and effective. To date, despite its effectiveness, there has little effort explore their robustness the impact of data poisoning attacks on performance. Can neighborhood‐based be easily fooled? this end, we shed light propose a novel attack framework, encoding purpose constraint against them. We first illustrate how calculate...
Three-dimensional (3D) ground-penetrating radar is an effective method for detecting internal crack damage in pavement structures. Inefficient manual interpretation of images and high personnel requirements have substantially restrained the generalization 3D radar. An improved Crack Unet model based on semantic segmentation proposed herein image processing. The experiment showed that MPA, MioU, accuracy were improved, it displayed better capacity task than current mainstream algorithms do,...
Under the circumstances of multi-VSG, virtual synchronous generator (VSG) control is liable to produce active power oscillation due absence mutual damping, which may lead fluctuations more acute that in turn cause instability problems. This study concentrates on analysis and optimization damping within multi-VSG systems. By developing a predictive model for VSG, strategy, incorporates both Model Predictive Control (MPC) consensus algorithm, proposed. The proposed method primarily employs MPC...
Vehicle re-identification (re-ID) is a challenging task in computer vision. Different vehicle identities belonging to the same model may have similar appearance with little inter-instance discrepancy, while one large intra-instance differences under different viewpoint and illumination. In this paper, we propose refined part learn an efficient feature embedding solve problem. from other methods, which directly obtain region for re-ID. The formulated through Grid Spatial Transformer Network...