A5102608392- Robotic Path Planning Algorithms
- Smart Grid Energy Management
- Robotic Locomotion and Control
- Microgrid Control and Optimization
- Control and Dynamics of Mobile Robots
- Context-Aware Activity Recognition Systems
- Age of Information Optimization
- Green IT and Sustainability
- Particle accelerators and beam dynamics
- Nuclear reactor physics and engineering
- Machine Learning in Bioinformatics
- Superconducting Materials and Applications
- Visual Attention and Saliency Detection
- Genetics, Bioinformatics, and Biomedical Research
- Power Systems and Technologies
- Electric Vehicles and Infrastructure
- Optimal Power Flow Distribution
- Advanced Neural Network Applications
- Smart Grid and Power Systems
- Generative Adversarial Networks and Image Synthesis
- Energy Load and Power Forecasting
- Electric Power System Optimization
- Biomedical Text Mining and Ontologies
- Advanced Battery Technologies Research
Duke University
2025
North China Electric Power University
2014
MinebeaMitsumi (Japan)
2005
Due to the short-term large-scale access of renewable energy and residential electric vehicles in communities, voltage limit distribution network will be exceeded, quality power supply seriously reduced. Therefore, this paper introduces mobile storage system (MESS), which effectively solves problem overvoltage caused by large number distributed sources household network. This proposes an optimal scheduling model for based on system. First, space-time transfer is established, cost MESS income...
Deep neural networks (DNNs) have demonstrated exceptional performance across various image segmentation tasks. However, the process of preparing datasets for training DNNs is both labor-intensive and costly, as it typically requires pixel-level annotations each object interest. To mitigate this challenge, alternative approaches such using weak labels (e.g., bounding boxes or scribbles) less precise (noisy) can be employed. Noisy are significantly quicker to generate, allowing more annotated...
It is conducive for the improvement of power system's reliability and economy to make scientific time-of-use (TOU) pricing peak, flat valley load, which guides customer have a good habit using energy, as well mechanism demand response (DR). To increase effectiveness TOU pricing, it necessary grasp laws by appropriate models improve accuracy analyzing customer's pricing. Firstly, this paper proposes load shifting adjustment coefficient current model; secondly, builds quantitative...
The radio-frequency quadrupole (RFQ) at the Facility for Rare Isotope Beams (FRIB) is a critical device to accelerate heavy ion beams from 12 keV/u 0.5 MeV/u state-of-the-art nuclear physics experiments. Efficient control of RFQ resonance frequency detuning still remains challenge because temperature-sensitive solely by cooling water system, exhibiting complicated transport delay and nonlinearity in heat transfer processes. In this work, we propose long-short term memory (LSTM)-based Koopman...
Distributed photovoltaics (DPVs) have been widely integrated into power systems due to their abundance, renewability and low cost, while the stochastic nature of DPVs imposes significant influence on hosting capacity (HC) DPVs. Here, electricity heat system (IEHS) is explored increase deployment in comparison by exploiting interaction between electric thermal energy. An MILP-based HC assessment model for IEHS proposed efficiently promote penetration level DPV generation distribution networks...
Mobile devices, especially smartphones, can support rich functions and have developed into indispensable tools in daily life. With the rise of generative AI services, smartphones potentially transform personalized assistants, anticipating user needs scheduling services accordingly. Predicting intents on reflecting anticipated activities based past interactions context, remains a pivotal step towards this vision. Existing research predominantly focuses specific domains, neglecting challenge...
Predicting genetic mutations from whole slide images is indispensable for cancer diagnosis. However, existing work training multiple binary classification models faces two challenges: (a) Training classifiers inefficient and would inevitably lead to a class imbalance problem. (b) The biological relationships among genes are overlooked, which limits the prediction performance. To tackle these challenges, we innovatively design Biological-knowledge enhanced PathGenomic multi-label Transformer...
Mobile devices, especially smartphones, can support rich functions and have developed into indispensable tools in daily life. With the rise of generative AI services, smartphones potentially transform personalized assistants, anticipating user needs scheduling services accordingly. Predicting intents on reflecting anticipated activities based past interactions context, remains a pivotal step towards this vision. Existing research predominantly focuses specific domains, neglecting challenge...
This paper describes the latest research progress of advanced distribution management system (ADMS) in networks area. Firstly, basic concepts, core and application functions, key technologies, main benefits ADMS are introduced. Further, several typical cases worldwide recent years Combined with its technical concept actual situation, we discussed analyzed multistage development route platform. Moreover, some suggestions for construction given.
This paper proposes a concept of initial minimum safety spacing (IMSS) to avoid the collision between robot and moving obstacles from any direction. A quick mtelligent control system based on IMSS is presented. In system, necessary deceleration steering for avoiding can be predicted using fuzzy inference, which only needs input one variable dangerous degree judged according IMSS. So, membership functions rules are very simple, calculation time obstacle avoidance reduced. The numerical...