A5100687298- Oceanographic and Atmospheric Processes
- Climate variability and models
- Tropical and Extratropical Cyclones Research
- Ocean Waves and Remote Sensing
- Coastal wetland ecosystem dynamics
- Meteorological Phenomena and Simulations
- Video Surveillance and Tracking Methods
- Marine and coastal ecosystems
- Atmospheric chemistry and aerosols
- Fluid Dynamics and Turbulent Flows
- Air Quality and Health Impacts
- Advanced Neural Network Applications
- Fire Detection and Safety Systems
- Hydrology and Watershed Management Studies
- Anomaly Detection Techniques and Applications
- Surface Treatment and Residual Stress
- Belt Conveyor Systems Engineering
- Artificial Immune Systems Applications
- Fluid Dynamics Simulations and Interactions
- IoT-based Smart Home Systems
- Evacuation and Crowd Dynamics
- Coastal and Marine Dynamics
- Nuclear reactor physics and engineering
- Modular Robots and Swarm Intelligence
- AI in cancer detection
Shandong Academy of Sciences
2012-2025
Qilu University of Technology
2019-2025
Shandong Institute of Automation
2012-2024
Shandong University
2023-2024
Ocean University of China
2012-2021
National Supercomputing Center in Wuxi
2017
National University of Singapore
2016
Abstract The overestimation of surface ozone concentration in low‐resolution global atmospheric chemistry and climate models has been a long‐standing issue. We first update the dry deposition scheme both high‐ (0.25°) (1°) Community Earth System Model (CESM) version 1.3 runs, by adding effects leaf area index correcting sunlit shaded fractions stomatal resistances. With this update, 5‐year‐long summer simulations (2015–2019) using CESM still exhibit substantial (by 6.0–16.2 ppbv) over U.S.,...
Using a three-dimensional (3D) hydrodynamic model, this study explored the seasonal transport structure in Pearl River Estuary and illustrated intrinsic connectivity under multiscale motions from Lagrangian perspective. Generally, surface residual current (U L ) is uniformly southwestward/southeastward summer/winter, with stronger intensity lower estuary. The bottom U features southeastward direction upper estuary northwest fluvial–tide interaction line advances northwestward summer winter,...
In this article, we use the Alexnet network in deep learning to determine whether lung CT images are infected with covid-19.First of all, data preprocessing stage, original image is scaled and normalized reduce noise interference. The batch operation training set test can increase speed;Secondly, build an eight-layer model, reasonable hyperparameters for each layer network, define loss function optimizer, processed train weight parameters model.Finally, three indicators accuracy, accuracy...
Nitrogen dioxide (NO2) is an important precursor of atmospheric aerosol. Forecasting urban NO2 concentration vital for effective control air pollution. This paper proposes a hybrid deep learning model predicting daily average concentrations on the next day, based pollutants, meteorological data, and historical data during 2014 to 2020 in five coastal cities Shandong peninsula, northern China. A random Forest (RF) algorithm was used select input variables reduce dimensionality trained by...
Introduction Currently, deep-learning-based prediction of Significant Wave Height (SWH) is mostly performed for a single location in the ocean or simply relies on factor (SF). Such approaches have disadvantage lacking spatial correlations dynamic complexity, leading to an inevitable growth error with time. Methods Here, attempting solution, we develop Multi-Factor (MF) data-driven 2D SWH model Bohai, Yellow, and East China Seas (BYECS). Our developed based multi-channel PredRNN algorithm...
Abstract. The Community Earth System Model (CESM) developed by the National Center for Atmospheric Research (NCAR) has been used worldwide climate studies. This study extends efforts of CESM development to include an online (i.e., in-core) ensemble coupled data assimilation system (CESM-ECDA) enhance CESM's capability predictability studies and prediction applications. CESM-ECDA consists atmospheric (ADA) component implemented in both finite-volume spectral-element dynamical cores ocean...
Mesoscale eddies can significantly influence the surface and deep layers of ocean, leading to diverse outcomes. Particularly, high-amplitude mesoscale possess vertical structures that reach thousands meters, carrying greater kinetic energy. Accurately detecting such is crucial for various applications, ranging from promoting marine economic development ensuring navigational safety. This study presents a cutting-edge algorithm eddy identification, leveraging multi-task learning multi-source...
Abstract This study investigates the seasonal variation in temperature induced by diurnal forcing eastern shelf seas of China (ESSC) using a high‐resolution Regional Ocean Modeling System model forced National Center for Environmental Prediction and Atmospheric Research re‐analysis data surface fluxes with both 6 h daily frequencies, respectively. The comparison between two experiments revealed modification forcing, which also increases mixed‐layer depth (MLD) August 26% reduces volume...
Abstract The impacts of diurnal atmospheric forcing on the summer salinity change in East China Sea are investigated using Regional Ocean Modeling System, forced by hourly and daily reanalysis wind insolation. differences between these two frequencies reveal a dipole pattern with positive deviation (1–2 psu) offshore Yangtze River estuary, negative (from −1 to −0.5 along Jiangsu Coast. Further dye tracking experiments confirm that strengthened northwestward longshore freshwater transport...
In order to improve the sorting efficiency of logistics and reduce fault tolerance rate, a three-wheel driven omnidirectional reconfigurable conveyor belt is designed in this paper. The we proposed composed number hexagonal modules, which each module has three wheels installed different directions. By adjusting speed omni-wheel, it possible move goods at given speed, adjust attitude angle process movement so as arbitrarily change shipping pose. Furthermore, algorithm optimal path planning...
In this paper, we propose an improved YOLOv5 pedestrian detection algorithm to solve the problems of target missing and low accuracy in ROS platform. By adding a small layer 160*160, method improves performance model effectively reduces false rate occluded pedestrians, especially heavily targets. order further improve accuracy, it fuses underlying features backbone network achieve path aggregation with multi-feature fusion. Furthermore, Soft-DIoU-NMS is used for post-detection processing...
In order to reduce radiation hazards, a mobile robot for detecting and grasping radioactive sources is designed. The based on the ROS platform, uses source detector locate moves source, mechanical gripper carried by body grab corresponding source. improve its stability adaptability, remote real-time control method used realize motion capture of detection robot. operates reducing environmental pollution open-air at same time avoiding damage staff.
Mesoscale eddies are common in global oceans, playing crucial roles ocean dynamics, circulation, and heat transport, their vertical structures can affect the water layers from tens to thousands of meters. In this study, we integrated sea surface height temperature data into deep learning methods study mesoscale eddy subsurface structure explore relationship between layers. introduce Dual_EddyNet, a algorithm designed invert eddies. Using algorithm, impact on inversion Furthermore, compare...
To realize the need for a push robot to move on fixed route in complex pasture environment, this paper proposes path-planning method that integrates improved A* algorithm and DWA. The segmented interpolation is used interpolate path points global planned by improve fitness demand of moving operation route. simulation results show compared with traditional DWA algorithm, fusion fitting distance reduced 21.36%, which can not only complete environment but also obstacle avoidance time, easier be...