A5080683301- Vibration Control and Rheological Fluids
- Vehicle Dynamics and Control Systems
- Advanced Thermoelectric Materials and Devices
- Hydraulic and Pneumatic Systems
- Thermal Radiation and Cooling Technologies
- Electric and Hybrid Vehicle Technologies
- Magnetic Bearings and Levitation Dynamics
- Advanced Battery Technologies Research
- Innovative Energy Harvesting Technologies
- Structural Engineering and Vibration Analysis
- Thermal properties of materials
- Seismic Performance and Analysis
- Advancements in Battery Materials
- Industrial Technology and Control Systems
- Mechanical Engineering and Vibrations Research
- Advanced Neural Network Applications
- Traffic control and management
- Autonomous Vehicle Technology and Safety
- Advanced Sensor and Control Systems
- Simulation and Modeling Applications
- Advanced Battery Materials and Technologies
- Natural Language Processing Techniques
- Heat Transfer and Optimization
- Electric Vehicles and Infrastructure
- Privacy-Preserving Technologies in Data
Jiangsu University
2016-2025
Jiangsu University of Science and Technology
2023-2025
Yangzhou University
2024-2025
Ministry of Agriculture and Rural Affairs
2025
Tianjin University
2025
Collaborative Innovation Center of Chemical Science and Engineering Tianjin
2025
Inner Mongolia University of Technology
2024
Boston University
2024
First Affiliated Hospital of Xi'an Jiaotong University
2024
China University of Mining and Technology
2023-2024
Abstract Accelerated conversion by catalysis is a promising way to inhibit shuttling of soluble polysulfides in lithium–sulfur (Li–S) batteries, but most the reported catalysts work only for one direction sulfur reaction (reduction or oxidation), which still not root solution since fast cycled use species finally realized. A bidirectional catalyst design, oxide–sulfide heterostructure, proposed accelerate both reduction and oxidation insoluble discharge products (e.g., Li 2 S), indicating...
Federated learning (FL) has recently attracted increasing attention from academia and industry, with the ultimate goal of achieving collaborative training under privacy communication constraints. Existing iterative model averaging based FL algorithms require a large number rounds to obtain well-performed due extremely unbalanced non-i.i.d data partitioning among different clients. Thus, we propose FedDM build global objective multiple local surrogate functions, which enables server gain more...
The active battery thermal management system is critical for the security of electric vehicles. In this article, a novel and control strategy based on thermoelectric cooling are proposed. A coupling model between cooler pack built by MATLAB/Simscape software. precision verified through experimental bench test, with maximal deviation 0.56 °C (the accuracy temperature sensor ±0.1 °C). Further, predictive (MPC) results, it elucidated that MPC has superiority over...
The automotive thermoelectric generator (ATEG) system represents a significant advancement in exhaust waste heat recovery for conventional or extended-range engines. This study aims to comprehensively assess the performance advantages of novel ATEG equipped with segmented converging exchanger. numerical method is used conduct simulation analysis. An experimental bench established verify accuracy model. results indicated that model highly accurate, maximum error 2.7% compared data....
The GA–CNFs–Ni modified separator endows the “double high” sulfur cathode (5–10 mg cm<sup>−2</sup>, 90%) with a stable reversible capacity and superior rate performance.
To improve the reliability of active electromagnetic suspension and reduce energy consumption, a hybrid that consists linear motor passive damper in parallel is proposed this paper. First, dynamic model established regeneration control systems are built. Thereafter, regeneration, ride comfort, driving safety taken as objects. The effect damping values on different objects studied, best determined. Passive for comparison, comparative simulation analysis conducted. Finally, bench test 1/4...
This paper proposes a novel differentiable architecture search method by formulating it into distribution learning problem. We treat the continuously relaxed mixing weight as random variables, modeled Dirichlet distribution. With recently developed pathwise derivatives, parameters can be easily optimized with gradient-based optimizer in an end-to-end manner. formulation improves generalization ability and induces stochasticity that naturally encourages exploration space. Furthermore, to...