A5100421971- Advanced Graph Neural Networks
- Electric Vehicles and Infrastructure
- Complex Network Analysis Techniques
- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- Adversarial Robustness in Machine Learning
- Text and Document Classification Technologies
- Advanced Aircraft Design and Technologies
- Sentiment Analysis and Opinion Mining
- Functional Brain Connectivity Studies
- Data Management and Algorithms
- Multimodal Machine Learning Applications
- Electric Power System Optimization
- Machine Learning and Algorithms
- Domain Adaptation and Few-Shot Learning
- Wind Energy Research and Development
- Advanced Image and Video Retrieval Techniques
- Artificial Immune Systems Applications
- Water resources management and optimization
- Water-Energy-Food Nexus Studies
- Smart Grid Energy Management
- Vehicle emissions and performance
- Natural Language Processing Techniques
- Topic Modeling
- Transportation and Mobility Innovations
Chinese Academy of Sciences
2023
Institute of Computing Technology
2023
University of Southern Denmark
2020-2022
National Grid (United Kingdom)
2021
UNSW Sydney
2021
South China University of Technology
2019
Changjiang Institute of Survey, Planning, Design and Research
2018
University of Bradford
2007
The passenger car sector accounts for nearly half of all transport-related emissions. Rapid decarbonization is therefore important but challenging, particularly in China, where, despite recent emission peak (by 2030) and carbon-neutrality 2060) pledges, ownership associated emissions are increasing. Successful reduction will require a rapid transition both technology (e.g., toward electrification) demand driving less). However, how to successfully deploy these twin strategies optimal...
Network embedding has recently attracted attention a lot since networks are widely used in various data mining applications. Attempting to break the limitations of pre-set meta-paths and non-global node learning existing models, we propose simple but effective framework for heterogeneous network by encoding original multi-type nodes relations directly self-supervised way. To be more specific, first learn relation-based embeddings global from neighbor properties under each relation type...
This paper proposes a new concept of Aviation-to-Grid (A2G) that utilises electric aircraft (EA) charging to provide flexibility the power grid.Smart EA system with battery swap method is developed using Photovoltaics (PV), gas turbine, and grid electricity.Hourly energy dispatch strategy produced based on mixed integer linear programming meet electrified aviation demand A2G frequency response grid.Case studies are conducted in 8 major UK airports considering seasonal flight schedules...
In 2015 the United Nations proposed Sustainable Development Goals (SDGs), a set of universal goals for meeting urgent environmental, political and economic challenges in world. Universities play an important role to support contribute SDGs mainly through education research. To evaluate contributions research, universities aim at relating their scientific publications SDGs, automatically quantify connectedness these detailed targets unique indicators under SDGs. this paper, we apply deep...
Selectivity estimation aims at estimating the number of database objects that satisfy a selection criterion. Answering this problem accurately and efficiently is essential to many applications, such as density estimation, outlier detection, query optimization, data integration. The especially challenging for large-scale high-dimensional due curse dimensionality, large variance selectivity across different queries, need make estimator consistent (i.e., non-decreasing in threshold). We propose...
Embedding learning in multiplex networks has drawn increasing attention recent years and achieved outstanding performance many downstream tasks. However, most existing network embedding methods either only focus on the structured information of graphs, rely human-annotated data, or mainly multi-layer GCNs to encode graphs at risk ill-posed spectral filters. Moreover, it is also challenging learn consensus embeddings for nodes across multiple views by inter-relationship among graphs. In this...
Complex networks represented as node adjacency matrices constrains the application of machine learning and parallel algorithms. To address this limitation, network embedding (i.e., graph representation) has been intensively studied to learn a fixed-length vector for each in an space, where properties original are preserved. Existing methods mainly focus on vectors preserve nodes proximity, i.e., next other space should also be closed but do not enforce algebraic statistical shared between...
Hydropower is a natural renewable energy with outstanding technological superiority in short-term operation. Short-term hydro generation operation (SHGO) requires not only lord dispatch single station, but also the coordination of hydraulic and electric connection for different stations. In this paper, coordinated peak shaving method cascade hydropower stations proposed. Firstly, regular pattern load regulation analyzed using characteristics power grid. The daily output should meet needs...