A5100744730- Speech and Audio Processing
- Advanced Adaptive Filtering Techniques
- Acoustic Wave Phenomena Research
- Multilevel Inverters and Converters
- Dynamics and Control of Mechanical Systems
- Image and Signal Denoising Methods
- Advanced DC-DC Converters
- Coal Combustion and Slurry Processing
- Geomechanics and Mining Engineering
- High voltage insulation and dielectric phenomena
- Biosensors and Analytical Detection
- Magnetic Bearings and Levitation Dynamics
- Nanomaterials for catalytic reactions
- Inertial Sensor and Navigation
- Coal Properties and Utilization
- Concrete and Cement Materials Research
- Power Systems and Renewable Energy
- Advanced Neural Network Applications
- Structural Behavior of Reinforced Concrete
- Wireless Power Transfer Systems
- Innovative concrete reinforcement materials
- Catalytic Processes in Materials Science
- Iterative Learning Control Systems
- Catalysis and Oxidation Reactions
- Tribology and Lubrication Engineering
North China Electric Power University
2025
Henan Polytechnic University
2024
South China University of Technology
2024
Institute of Acoustics
2019-2023
Chinese Academy of Sciences
2023
Foshan University
2018
Nanjing University
2008
Abstract Thanks to the rapid development of image processing technology, efficiency and accuracy power inspection have been enhanced through deep learning techniques. However, during on-site inspections, complexity background images composite insulators often makes it difficult directly extract key features for accurately assessing hydrophobicity levels. Moreover, considering real-time requirements insulator detection in practical operations, model must be lightweight speed up process. To...
ABSTRACT In the field of power equipment inspection, aging condition composite insulators is often determined by detection water repellency. However, existing methods are difficult to effectively extract repellency level features in complex background, and it meet real‐time requirements. Therefore, this paper proposes a MDC‐YOLO algorithm for classification insulators. The combines efficient convolutional network (EfficientNetV2), deformable attention mechanism (DAttention), lightweight...
Abstract A global quasi‐sliding mode control ( GQSMC ) scheme is introduced to ensure zigzag motion with a smaller bound than that offered by Gao's method and provide disturbance rejection throughout the entire response in discrete time. The design of an augmented forcing function followed three conditions time extended from sliding GSMC continuous Furthermore, we adopt switching gain, which auto‐tuned as surface s(k), such chattering phenomena can be considerably alleviated during...
When implementing active noise control (ANC) systems, the filtered-x LMS (FXLMS) algorithm is often used, where secondary path transfer function between output of adaptive filter and error sensor has to be modeled. To model precisely, estimation should sufficient order. However, too large an order increases computation load FXLMS algorithm. In this letter, relationship modeling phase analyzed.
The centralized control for multi-channel active noise (ANC) systems usually cost considerable processing power due to the transfer functions between a large number of loudspeakers and error microphones; while decentralized has increased risk global instability. Distribution controller network could save computational burden maintaining satisfactory performance. A lot ANC algorithms employing diffusion have been proposed recently. communication within networks introduces estimation bias each...
Abstract The variable inductor integrated LLC (VI‐LLC) converter is able to operate at constant switching frequency and avoid wide range thus emerged in recent years. However, VI‐LLC suffers two reliability problems. First, coupling between the main winding control of (VI) causes negative effects on converter, which are few considered literature. Second, parameter fluctuation (PF) resonant elements has operation gain causing unreliable voltage deviation. To solve above problems, this paper,...
Virtual Synchronous Generator uses dead-beat predictive control in the voltage and current loop to improve dynamic response speed of system, but presence zero-order hold computation time can lead system instability. To address this issue, paper adopts an improved strategy for VSG. This is achieved by linear interpolation forward predict voltage-current reference values reduce impact delay, while introducing error compensation coefficients enhance stability inner The selection based on...