A5030296824- Advanced Memory and Neural Computing
- Mobile Agent-Based Network Management
- Service-Oriented Architecture and Web Services
- Advanced Computational Techniques and Applications
- Neural dynamics and brain function
- Advanced Algorithms and Applications
- Smart Grid and Power Systems
- Photoreceptor and optogenetics research
- Power Systems and Technologies
- HVDC Systems and Fault Protection
- IoT and Edge/Fog Computing
- Microgrid Control and Optimization
- Big Data and Digital Economy
- High-Voltage Power Transmission Systems
- IPv6, Mobility, Handover, Networks, Security
- Smart Grid Security and Resilience
- Software-Defined Networks and 5G
- Industrial Vision Systems and Defect Detection
- Ovarian cancer diagnosis and treatment
- Ferroelectric and Negative Capacitance Devices
- Computational Drug Discovery Methods
- Technology and Security Systems
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Optical measurement and interference techniques
- Energy Load and Power Forecasting
Changsha Medical University
2024-2025
Tangshan Gongren Hospital
2025
Huazhong University of Science and Technology
2015-2024
South China University of Technology
2024
University of California, Santa Cruz
2020-2024
Taizhou University
2024
Lanzhou University
2024
Ministry of Agriculture and Rural Affairs
2024
University of Bonn
2024
State Grid Corporation of China (China)
2018-2024
This paper proposes an improved coordinate direct power control (DPC) strategy for the doubly fed induction generator (DFIG) and grid side converter (GSC) of a wind generation system under unbalanced network conditions. Two DPC schemes DFIG GSC are presented, respectively. The is to eliminate torque stator reactive pulsations, while compensate pulsations active power. In order provide enhanced performance overall system, resonant controllers tuned at fundamental double frequency applied in...
Spiking neural networks and neuromorphic hardware platforms that simulate neuronal dynamics are getting wide attention being applied to many relevant problems using Machine Learning. Despite a well-established mathematical foundation for dynamics, there exists numerous software solutions stacks whose variability makes it difficult reproduce findings. Here, we establish common reference frame computations in digital systems, titled Neuromorphic Intermediate Representation (NIR). NIR defines...
We present and experimentally validate two minimal compact memristive models for spiking neuronal signal generation using commercially available low-cost components. The first neuron model is called the Memristive Integrate-and-Fire (MIF) model, signaling with voltage levels: spike-peak, rest-potential. second MIF2 also presented, which promotes local adaptation by accounting a third refractory level during hyperpolarization. show both are in terms of number circuit elements integration...
Background The incidence of Chronic Inflammatory Airway Diseases (CIAD) has been steadily increasing, making it a significant contributor to the global disease burden. Additionally, risk airway diseases in elderly women continues rise each year, with nutritional factors playing crucial role progression CIAD. Geriatric Nutritional Risk Index (GNRI) is novel tool for assessing individual status. This study aims assess relationship between GNRI and all-cause cardiovascular mortality CIAD,...
We present MEMprop, the adoption of gradient-based learning to train fully memristive spiking neural networks (MSNNs). Our approach harnesses intrinsic device dynamics trigger naturally arising voltage spikes. These spikes emitted by are analog in nature, and thus differentiable, which eliminates need for surrogate gradient methods that prevalent network (SNN) literature. Memristive typically either integrate memristors as synapses map offline-trained networks, or otherwise rely on...
Background Neutrophil-to-lymphocyte ratio (NLR) is considered a biomarker of systemic inflammation and immune activation. However, its relationship with the risk mortality in patients chronic obstructive pulmonary disease (COPD) remains unclear. This study aimed to investigate association between NLR all-cause cardiovascular COPD. Methods Data were collected from National Health Nutrition Examination Survey (NHANES) January 1999 December 2018. The calculation method involves dividing...
Identifying influential nodes in complex networks is of both theoretical and practical importance. Existing methods identify based on their positions the network assume that are homogeneous. However, node heterogeneity (i.e., different attributes such as interest, energy, age, so on) ubiquitously exists needs to be taken into consideration. In this paper, we conduct an investigation propose a graph signal processing centrality (GSPC) method considering topology. We first evaluate our GSPC...
Due to the complex environment of coal mine, accidents can occur at any time and often result in partial or total evacuation mine personnel could loss lives. Therefore, it is important necessary detect generate a corresponding alarming disposal time. This paper proposed real-time event detecting processing approach for safety using wireless sensor network. Firstly, we introduce events model, offer fully customizable policies selection consumption, also describe state-automata-based detection...
This paper introduces a dual-layer zero trust architecture (ZTA) to enhance 5G vertical industry multi-access edge computing (MEC) application (APP) access control security. In the ZTA, policy engine is deployed at core network, which evaluates value of UE MEC APP service. User's network layer behavior and are jointly evaluated engine. Protocol interactive procedures designed realize proposed framework in real 3GPP-defined network. The exposure function open interfaces leveraged mobile...
It has been demonstrated that aberrant androgen receptor (AR) signaling contributes to the pathogenesis of prostate cancer (PCa). To date, most efficacious strategy for treatment PCa remains target AR axis. However, numerous patients still face issue overtreatment or undertreatment. The establishment a precise risk prediction model is urgently needed distinguish with high-risk and select appropriate modalities.
已有的聚类集算法基本上都是非监督聚类集成算法,这样不能利用已知信息,使得聚类集成的准确性、鲁棒性和稳定性降低。把半监督学习和聚类集成结合起来,设计半监督聚类集成模型来克服这些缺点。主要工作包括:第一,设计了基于贝叶斯网络的半监督聚类集成(semi-supervised cluster ensemble,简称SCE)模型,并对模型用变分法进行了推理求解;第二,在此基础上,给出了EM(expectation maximization)框架下的具体算法;第三,从UCI(University of Califor;The existing algorithms are mostly unsupervised a ensemble, which cannot take advantages known information datasets. As result, the precision, robustness, and stability ensemble degraded. To conquer these disadvantages, semi-supervised...
We present a fully memristive spiking neural network (MSNN) consisting of physically-realizable neurons and synapses to implement an unsupervised Spike Timing Dependent Plasticity (STDP) learning rule. The system is in that both neuronal synaptic dynamics can be realized by using memristors. neuron implemented the SPICE-level integrate-and-fire (MIF) model, which consists minimal number circuit elements necessary achieve distinct depolarization, hyperpolarization, repolarization voltage...
Power systems have developed significantly because of the increasing share renewable energy sources (RESs). Despite advantages, they also bring inevitable challenges to power system stability, especially under extreme fault conditions. This paper presents a practical active support control strategy for RESs grid The proof process is taken in an AC‐DC hybrid integrated with large capacity PV stations and wind farms. on‐site engineering test results reflect that potential risks operation...
Spiking neural networks and neuromorphic hardware platforms that emulate dynamics are slowly gaining momentum entering main-stream usage. Despite a well-established mathematical foundation for dynamics, the implementation details vary greatly across different platforms. Correspondingly, there plethora of software implementations with their own unique technology stacks. Consequently, systems typically diverge from expected computational model, which challenges reproducibility reliability...
We are developing a hardware platform based on FPGA with highly parallel neuromorphic computing architecture that can be utilized in drone radiation detector for anomaly detection. Driven by national security threat landscape, there is need detectors readout circuits low size, weight, and power consumption capable of implementing realtime machine learning. Anomalous radioactive source detection task has the limitation far detecting distance, limited dwell time, weak activity, which...
New features like high penetration level, low inertia, and weak damping are decreasing voltage support capacity in the power system as it sees large-scale renewable energy source (RES) integration continuously. The reasonable solution to this challenge is developing an additional active regulation function for converters (VSCs). This study proposes a droop-based control strategy VSCs RES-integrated systems make continue operate regulate at point of common coupling (PCC) during fault without...
Fibroblast growth factors (FGFs) are required for the specification and formation of epibranchial placodes, which give rise to distal part cranial sensory ganglia. However, it remains unclear whether FGFs play a role in regulating neurite outgrowth placode-derived ganglia during further development. Previous studies have shown that factor 8 (FGF8) promotes from statoacoustic ganglion vitro. these did not distinguish between neural crest- components In this study, we focused on petrosal...