A5101502125- Rock Mechanics and Modeling
- Landslides and related hazards
- Privacy-Preserving Technologies in Data
- Power Systems and Renewable Energy
- Smart Grid and Power Systems
- Wind Turbine Control Systems
- Geotechnical Engineering and Analysis
- HVDC Systems and Fault Protection
- Seismic Waves and Analysis
- Microgrid Control and Optimization
- earthquake and tectonic studies
- Electric Power System Optimization
- Seismic Imaging and Inversion Techniques
- Stochastic Gradient Optimization Techniques
- Dam Engineering and Safety
- Adaptive optics and wavefront sensing
- Cryptography and Data Security
- Seismology and Earthquake Studies
- Power System Reliability and Maintenance
- Advanced Algorithms and Applications
- Geophysical Methods and Applications
- Recommender Systems and Techniques
- Geothermal Energy Systems and Applications
- Machine Learning and ELM
- Optical Polarization and Ellipsometry
State Grid Corporation of China (China)
2024
Jilin University
2023
Dalian University of Technology
2020-2023
Tiangong University
2023
China University of Petroleum, East China
2023
Qingdao University of Technology
2023
Changchun University of Science and Technology
2023
Taiyuan Heavy Industry (China)
2022
Taiyuan University of Technology
2021
Shanghai Technical Institute of Electronics & Information
2019-2020
Cross-device Federated Learning (FL) is a distributed learning paradigm with several challenges that differentiate it from traditional learning, variability in the system characteristics on each device, and millions of clients coordinating central server being primary ones. Most FL systems described literature are synchronous - they perform synchronized aggregation model updates individual clients. Scaling challenging since increasing number training parallel leads to diminishing returns...
Recent years have witnessed the increasing risk of subsea gas leaks with development offshore exploration, which poses a potential threat to human life, corporate assets, and environment. The optical imaging-based monitoring approach has become widespread in field underwater leakage, but shortcomings huge labor costs severe false alarms exist due related operators’ operation judgment. This study aimed develop an advanced computer vision-based achieve automatic real-time leaks. A comparison...
In order to mitigate the risk of irreversible drowning injuries, this study introduces an enhanced YOLOv5 algorithm aimed at improving efficacy indoor swimming pool detection and facilitating timely rescue endangered individuals. To simulate positions accurately, four swimmers were deliberately chosen observed, with monitoring conducted by drones flying above pool. The was approved ethics committee our institution, registration number 2022024. images captured underwent a meticulous...
Due to the different geological conditions and construction methods associated with projects, rockbursts in deep-buried tunnels often present precursor characteristics, bringing major challenges early warning of rockbursts. To adapt complexity engineering, it is necessary review latest advancements rockburst discuss general methods. In this article, first, microseismic monitoring localization applicable under tunneling are reviewed. Based on engineering examples research progress, evolution...
In adapting to the double-high development trend of high-voltage direct current (HVDC) receiving-end power systems and solving optimization problems in emergency frequency control (EFC) supporting virtual plants (VPPs) large-scale systems, a parameter estimation method for VPP response model based on successive linear programming (SLP) is proposed. First, “centralized/decentralized” hierarchical architecture participation EFC designed. Second, characteristics multiple flexible resources are...
Deepening our understanding of temperature and stress evolution in high-temperature tunnels is indispensable for tunnel support associated disaster prevention as the rock remarkably high hot dry (HDR) utilization similar engineering. In this paper, we established a two-dimensional thermal–mechanical coupling model through RFPA2D-thermal, by which field surrounding with without thermal insulation layer (TIL) were studied, followed cracks. The sensitivity analysis TIL airflow factors then...
Tunnel excavation inevitably causes surface deformation. In urban areas, deformation could lead to the of surrounding buildings, which may cause damage communities when accumulated a certain extent. However, current construction organization and management mainly rely on on-site measurements, there is still lack reliable prediction methods. Here, we proposed an effective evaluation method for frame building based stochastic medium theory equivalent beam theory. This effectively evaluate...
The microseismic signals in the coal minefield are very complex because of its special environment with a large number blast vibration signals, and how to effectively identify is still big problem. S transform (ST) Manifold Learning (ML) methods introduced extract characteristics Gaussian Mixture Model based on improved Bee Colony optimization algorithm (IBC‐GMM) established accurately. Firstly, time‐frequency mine extracted by ST analysis. It found that there obvious differences between...
Federated learning (FL) has emerged as an effective approach to address consumer privacy needs. FL been successfully applied certain machine tasks, such training smart keyboard models and keyword spotting. Despite FL's initial success, many important deep use cases, ranking recommendation have limited from on-device learning. One of the key challenges faced by practical adoption for DL-based is prohibitive resource requirements that cannot be satisfied modern mobile systems. We propose...
The location algorithm is the core of microseismic monitoring, and results directly influence effectiveness subsequent source analysis. We introduced double-difference (DD) method in seismic into monitoring. built a travel time equation layered velocity model using ray-tracing theory analyzed waveform cross-correlation data generalized (GCC) technique. improved particle swarm was successfully conducted to improve randomness initial location. then adopted DD precisely locate secondary...
In the original publication [...]
Cross-device Federated Analytics (FA) is a distributed computation paradigm designed to answer analytics queries about and derive insights from data held locally on users' devices. On-device computations combined with other privacy security measures ensure that only minimal transmitted off-device, achieving high standard of protection. Despite FA's broad relevance, the applicability existing FA systems limited by compromised accuracy; lack flexibility for analytics; an inability scale...
Abstract In this paper, evolutionary extreme learning machine (E‐ELM) is first introduced for RF power amplifiers (PAs) behavioral modeling. This approach combined differential evolution (DE) and (ELM) to effectively solve the problem that more neurons of hidden layer are required, repeated trials necessary in modeling PAs by conventional ELM. As revealed practices on Class‐AB Class‐E PAs, fewer used than condition Meanwhile, it found ELM's unstable generalization ability also significantly...