Browsing by Author "Chen, Gang"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Electrochemical mechanisms of an advanced low-temperature fuel cell with a SrTiO 3 electrolyte
    (ROYAL SOC CHEMISTRY, 2019-04-28) Chen, Gang; Liu, Hailiang; He, Yang; Zhang, Linlin; Asghar, Muhammad Imran; Geng, Shujiang; Lund, Peter D.; Department of Applied Physics; New Energy Technologies; Northeastern University China
    The electrochemical mechanisms and performance of a symmetrical low-temperature SOFC with a single oxide as the electrolyte are investigated here. The fuel cell has a layered Ni foam-Ni 0.8 Co 0.15 Al 0.05 LiO 2 (NCAL)/SrTiO 3 (STO)/NCAL-Ni foam structure. A 0.8 mm thick layer of STO is used as the electrolyte and NCAL-coated nickel foam is used as the electrode on both sides of the cell. The maximum power densities of the cell were 0.31, 0.44, and 0.62 W cm −2 in a H 2 /air atmosphere at 450, 500, and 550 °C, respectively. The corresponding ionic conductivities of the STO electrolyte were 0.16, 0.21, and 0.24 S cm −1 . Ion filtration experiments with densified Gd-doped CeO 2 /STO and SrCe 0.95 Y 0.05 O 3−δ /STO double layer electrolytes indicated that both oxygen ions and protons act as charge carriers in the STO electrolyte. XPS, TGA, and HRTEM analyses indicate that lithium carbonate, which originates from the NCAL, coats the STO electrolyte and forms a core-shell structure in the fuel cell test atmosphere. Lithium carbonate between the surface and interface of the STO particles may provide a pathway for oxygen ion and proton conduction.
  • No Thumbnail Available
    Item
    Investigation of the fracture mechanism of level ice with extended finite element method
    (PERGAMON-ELSEVIER SCIENCE LTD, 2022-09-15) Xu, Ying; Wu, Jiameng; Li, Pingshu; Kujala, Pentti; Hu, Zhiqiang; Chen, Gang; Department of Mechanical Engineering; Marine Technology; Marine Design and Research Institute of China; Newcastle University
    This paper investigates the fracture mechanism of level ice based on the extended finite element method by simulating collision scenarios between ice and a rigid ship structure. It is found the collision velocity and structure inclination affect the fracture mode through changing the deformation and stress distribution of the level ice. The overall response of the level ice is simulated with the transversely isotropic material model and cohesive zone model. The numerical model is verified with the data from a field test, which shows that the obtained ice load and size of the broken ices from numerical method are well consistent with the tested data. Two fracture modes of the level ice, bending and splitting, appear in the simulated cases. The bending crack is found to emerge from the top surface of the level ice and expand along the circumferential direction, and the splitting crack initiates at the bottom edge of the level ice and expands along the radial direction. Deformation and multiple stresses of level ice are analyzed, showing that the initial cracks for both fracture modes are related to the local tensile failure, and the location of the maximum tensile hydrostatic stress always coincides with the initial crack.
  • Thumbnail Image
    Item
    Mechanism for Major Improvement in SOFC Electrolyte Conductivity When Using Lithium Compounds as Anode
    (AMERICAN CHEMICAL SOCIETY, 2020-05-26) He, Yang; Chen, Gang; Zhang, Xuebai; Zhang, Linlin; Yang, Di; Asghar, Muhammad Imran; Geng, Shujiang; Lund, Peter D.; Department of Applied Physics; New Energy Technologies; Northeastern University China
    Recent studies indicate that the electrolyte ionic conductivity in an SOFC can be considerably increased by using lithium compounds as the electrode. We found that the ionic conductivity of Gd0.1Ce0.9O1.95 electrolyte in the Ni0.8Co0.15Al0.05LiO2 anode cell was 10.1 mS/cm only at 550 °C without H2, but it increased to 44.6 mS/cm after feeding H2 to the anode. It was found that LiOH/Li2CO3 moved into the GDC electrolyte from the NCAL anode and formed a three-phase composite electrolyte. A space charge region with a high oxygen vacancy concentration is formed around the interface of LiOH/Li2CO3 and GDC, which increases the ionic conductivity.
  • Thumbnail Image
    Item
    MSRL: Distributed Reinforcement Learning with Dataflow Fragments
    (2023) Zhu, Huanzhou; Zhao, Bo; Chen, Gang; Chen, Weifeng; Chen, Yijie; Shi, Liang; Yang, Yaodong; Pietzuch, Peter; Chen, Lei; Imperial College London; Department of Computer Science; Huawei Technologies Co., Ltd.; Peking University; Hong Kong University of Science and Technology
    A wide range of reinforcement learning (RL) algorithms have been proposed, in which agents learn from interactions with a simulated environment. Executing such RL training loops is computationally expensive, but current RL systems fail to support the training loops of different RL algorithms efficiently on GPU clusters: they either hard-code algorithm-specific strategies for parallelization and distribution; or they accelerate only parts of the computation on GPUs (e.g., DNN policy updates). We observe that current systems lack an abstraction that decouples the definition of an RL algorithm from its strategy for distributed execution. We describe MSRL, a distributed RL training system that uses the new abstraction of a fragmented dataflow graph (FDG) to execute RL algorithms in a flexible way. An FDG is a heterogenous dataflow representation of an RL algorithm, which maps functions from the RL training loop to independent parallel dataflow fragments. Fragments account for the diverse nature of RL algorithms: each fragment can execute on a different device through a low-level dataflow implementation, e.g., an operator graph of a DNN engine, a CUDA GPU kernel, or a multi-threaded CPU process. At deployment time, a distribution policy governs how fragments are mapped to devices, without requiring changes to the RL algorithm implementation. Our experiments show that MSRL exposes trade-offs between different execution strategies, while surpassing the performance of existing RL systems with fixed execution strategies.
  • No Thumbnail Available
    Item
    Numerical simulation of level ice impact on landing craft bow considering the transverse isotropy of Baltic Sea ice based on XFEM
    (Elsevier Ltd., 2020-05-01) Xu, Ying; Kujala, Pentti; Hu, Zhiqiang; Li, Fang; Chen, Gang; Department of Mechanical Engineering; Marine Technology; Newcastle University; Shanghai Jiao Tong University
    Ice bending is a major failure mechanism of level ice when ships and marine structures interact with level ice. This paper aims to investigate the ice bending and ice load when level ice collides on ships and marine structures using numerical simulation method, and compare the numerical results with field test. The fracture of ice is simulated with extended finite element method (XFEM), and cohesive zone concept is used to describe the crack propagation. In order to consider the characteristics of S2 columnar ice, a transversely isotropic elastic material model is used for the ice bulk elements, and a transversely isotropic Tsai-Wu failure criterion is adopted to predict the initiation of cracks. A well-controlled field test of a landing craft bow colliding with level ice in Baltic Sea is simulated to verify the numerical scheme. The ice plate's continuous deformation, crack initiation and crack propagation at different impact velocities and angles are simulated and the results are discussed. In the simulation, the bending crack emerges at the midline of the top surface of ice plate, then propagates towards free boundary, and finally a circumferential crack forms. It is found that with the impact velocity increases, the bending load increases and the fracture size (perpendicular distance from the crack to the contact edge) decreases. And as the angle between the landing craft bow and vertical direction increases, the bending load and the fracture size decrease. The simulated results corresponds well with the field test. The competition between the circumferential crack and radial crack is also found in the simulation and will be discussed in this paper. The results show that this method well simulates the bending of level ice and predict the ice load, and provides a good approach for investigating the mechanism of different forms of level ice fracture.