A5101961495- Electric Power System Optimization
- Microgrid Control and Optimization
- Optimal Power Flow Distribution
- Integrated Energy Systems Optimization
- Smart Grid Energy Management
- Electric Vehicles and Infrastructure
- Advanced Wireless Communication Techniques
- Advanced Measurement and Detection Methods
- Cooperative Communication and Network Coding
- Sparse and Compressive Sensing Techniques
- Advanced Vision and Imaging
- Adaptive Dynamic Programming Control
- Stability and Control of Uncertain Systems
- Neuroinflammation and Neurodegeneration Mechanisms
- Advanced Nanomaterials in Catalysis
- Error Correcting Code Techniques
- Advanced Image Processing Techniques
- Infrared Target Detection Methodologies
- Energy Load and Power Forecasting
- Power Systems and Renewable Energy
- Nanoparticle-Based Drug Delivery
- Recommender Systems and Techniques
- Caching and Content Delivery
- Smart Grid and Power Systems
- Image Processing Techniques and Applications
Guangdong University of Technology
2023-2024
South China University of Technology
2019-2023
China Southern Power Grid (China)
2023
Changzhou University
2022
University of Electronic Science and Technology of China
2021
Foshan University
2019
Xiamen University
2005
In this paper, we summarize the 1st NTIRE challenge on stereo image super-resolution (restoration of rich details in a pair low-resolution images) with focus new solutions and results. This has 1 track aiming at problem under standard bicubic degradation. total, 238 participants were successfully registered, 21 teams competed final testing phase. Among those participants, 20 submitted results PSNR (RGB) scores better than baseline. establishes benchmark for SR.
Abstract T cells play a basic and key role in immunotherapy against solid tumors, efficiently recruiting them into neoplastic foci sustaining long‐term effector function are consistent goals that remain critical challenge. Here, an injectable alginate‐based hydrogel with abundant β‐cyclodextrin (ALG‐βCD) sites is developed intratumorally injected to recruit CCR9 + CD8 (a subset of robust antitumor activity) via the trapped chemokine CCL25. In meantime, intravenously adamantane‐decorated...
Due to the increasing penetration of renewable energy source, power system is facing significant uncertainties. How fully consider those uncertainties in dynamic economic dispatch (DED) has become a crucial problem safe and operation system. Reinforcement learning (RL) based approaches can provide policy response uncertainty. However, current RL uses traditional Euclidean data representation, which greatly reduces scalability computational efficiency algorithms. To address such obstacle,...
Coordination of networked microgrids (MGs) offers a promising solution to utilize distributed resources flexibilities and accommodate renewable energy. This paper studies real-time coordination distribution system operation (DSO) MGs considering multivariate uncertainty. Current researches suffer from inadaptability dynamic uncertainty, extensive iterations, dependence on prediction. To fill these gaps, A novel multi-agent learning based stochastic programming (MASDP) is proposed obtain the...
With the construction of a new power system and proposal double carbon goal, operation data are growing explosively, optimization dispatching is becoming more complex. Relying on traditional pure manual difficult to meet needs. The emerging knowledge graph technology in field artificial intelligence one effective methods solve this problem. Because topological structure itself consistent with relational theory, through establishment relevant graph, real operating state can be restored...
Electric vehicle (EV) charging, and discharging have an important reactive power support capability for the distribution network, which a optimization method considering EV dis-/charging is proposed. Firstly, principle of charger's regulation analysed, charger state constraint model established to satisfy constraints on factor angle by modifying battery capacity target users. Secondly, quantification user subsidies proposed establish network-side user-side demand indicators respectively,...
In this paper, the performance of non-binary low-density parity-check (LDPC) codes is investigated with high rates (R > 2/3), and small block lengths (N < 5000 bits) over noise bursts channels, compared that Reed-Solomon (RS) codes. Meanwhile, a new optimized encoding scheme LDPC presented, decoding algorithm reduced-complexity introduced. The simulation results indicate are very effective against bursts, about 2.7 dB superior to RS at frame error rate (FER) FER=10/sup -4/, even when burst...
In this paper, we consider the clustered multi-task learning (MTL) problem with partial observations and develop a diffusion least-mean-square (LMS) algorithm distributed cluster-wise sampling strategy. The proposed converges at steady-state measurements observed only subset of vertices, instead entire graph, without significant loss performance. We analyze performance further devise tractable cost function respect to probability based on an approximate network Mean-Square-Deviation (MSD)...
Benefiting from the progress of power electronics technology, distributed generation technology is developing rapidly. Since micro grids cannot rely on traditional multi-time scale control strategies to ensure high-quality frequency stability and economic dispatch in same time scale, this paper proposes an extreme dynamic programming algorithm. The proposed algorithm takes adaptive as framework, learning machine a kernel evaluation module, model implementation module new prediction module....
Coordinated dispatch of integrated electricity and thermal system (IETS) provides extra operation flexibility which is further improved by integration electrical storages. However, the problem non-convexity multiple uncertainties hinder its optimal real-time dispatch. This paper proposes an approximate dynamic programming with imitation learning (ADP-IL) based policy for IETS storages, computationally efficient adaptive to while satisfying complex networks constraints. First, reformulated as...
This paper proposes a novel approach for dynamic economic dispatch (DED) of distribution networks, based on graph-generative adversarial network (Graph-GAN) assisted inverse reinforcement learning (IRL) with human knowledge transfer via demonstration. Firstly, the proposed method utilizes graph convolutional (GCN) to capture complex and nonlinear relationships between decision system state. Secondly, GAN-based is imitate reward function from expert demonstration data, which avoids need...
This paper proposes a novel multi-searcher optimization (MSO) algorithm for the optimal energy dispatch (OED) of combined heat and power-thermal-wind-photovoltaic systems. The available power wind turbine (WT) units photovoltaic (PV) is approximated with probability density functions speed solar irradiance, respectively. chaos theory used to implement wide global search, which can effectively avoid low-quality local optimum OED. Besides, double-layer searcher designed guarantee fast...
Due to the fast development of new-type power systems, grid is facing increasing uncertainties brought by high penetration distributed generations. How improve decision quality economic dispatch under such conditions becomes a very crucial task. Therefore, novel graph reinforcement learning (GRL) approach for dynamic renewable energy proposed. Compared with other method, graph-based representation system state adopted. Thus, implicit correlations considering topology can be more effectively...
Monitoring and measurement on footing resistance of transmission towers is an important safeguard to towers. This paper introduces a frequency conversion technology that greatly increasing efficiency accuracy The study analyses the device's hardware software systems, application result shows device helps easier more accurate monitor grounding resistance.
The spotlight source control system includes a single mode system. A controller controls 32 points through video signal to differential achieve point control, and arbitrarily sets the white balance of certain menu; main auxiliary machine online system, are connected by DMX signals, which can realize 1000 cascades (that is, total 32,000 cascade be realized), comes with variety effect modes, you freely change mode, connection so on. has wide range applications in landscaping urban night scenes.