Search NASA⌕ Search

Engineering topics

Chen, Zhe

Publications and source records attributed to Chen, Zhe.

28 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

In Situ Identification of NNH and N 2 H 2 by Using Molecular-Beam Mass Spectrometry in Plasma-Assisted Catalysis for NH 3 Synthesis

In this work, ammonia synthesis at 533 K and atmospheric pressure was investigated in a coaxial dielectric barrier discharge (DBD) plasma reactor without packing and with porous γ-Al 2 O 3 , 5 wt % Ru/γ-Al 2 O 3 , or 5 wt % Co/γ-Al 2 O 3 catalyst particles. Gas-phase species were monitored in situ using an electron impact molecular-beam mass spectrometer (EI-MBMS). Gas-phase species NNH and N 2 H 2 were first identified under common conditions of plasma-assisted ammonia synthesis and were present at levels comparable to that of NH 3 in the plasma discharge. Concentrations of NNH, N 2 H 2 , and NH in a reactor packed with γ-Al 2 O 3 or other particles were lower than those observed in an empty reactor, while the concentration of NH 3 increased. These observations point to the importance of NNH and N 2 H 2 in plasma-assisted surface reactions in ammonia synthesis. Reaction pathways of direct adsorption of gas-phase NNH and N 2 H 2 on solid surfaces and subsequent reactions were proposed. This study demonstrated that in situ identification of gas-phase species via EI-MBMS provides a powerful approach to study the kinetics of plasmaassisted catalysis.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Model-Free Voltage Control of Active Distribution System with PVs Using Surrogate Model-Based Deep Reinforcement Learning

Accurate knowledge of the distribution system topology and parameters is required to achieve good voltage control performance, but this is difficult to obtain in practice. This paper proposes a physical-model-free voltage control method based on a surrogate-model-enabled deep reinforcement learning approach. Specifically, a surrogate model is trained in a supervised manner using the recorded limited number of historical data to learn the relationship between the power injections and voltage fluctuations of each node. Then, the deep reinforcement learning algorithm is applied to learn an optimal control strategy from the experiences obtained by continuous interactions with the surrogate model. The proposed method can achieve physical-model-free control of unbalanced distribution network and inform real-time decisions to deal with fast voltage fluctuations caused by the rapid variation of PV generation. Simulation results on an unbalance IEEE 123-bus system show that the proposed method can achieve similar performance as that of perfect physical-model-based approaches while being advantageous over other traditional methods.

active distribution network↗
DOE OSTI2021

Plasma-assisted catalysis for ammonia synthesis in a dielectric barrier discharge reactor: key surface reaction steps and potential causes of low energy yield

Ammonia synthesis experiments were carried out in a coaxial dielectric barrier discharge (DBD) reactor packed with several different supports and metal catalysts. There was a marked increase in the reaction rate, over that obtained in an empty DBD plasma reactor, upon introduction of a packed bed of γ-Al 2 O 3 , Ru/γ-Al 2 O 3 , or SiO 2 particles. The difference in the reaction rates over γ-Al 2 O 3 and Ru/γ-Al 2 O 3 was minimal. Complementary zero-dimensional plasma kinetic model analysis was also performed using inputs from experimental data. This kinetic analysis allowed for gas phase reactions, Eley–Rideal (E–R) reactions, and direct adsorption of radical species on the γ-Al 2 O 3 surface. On the metal surface, dissociative adsorption of N 2 and H 2 , and Langmuir–Hinshelwood reactions were also included. This analysis revealed that, under the conditions of our experiments, ammonia synthesis proceeds principally by the formation of reactive radicals in the gas phase, which then adsorb and participate in E–R reactions on both the metal and support material surfaces. Furthermore, this finding illustrates a challenge for substantially increasing the energy yield for plasma-assisted ammonia synthesis in typical DBD reactors containing packed catalyst beads.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

A New Impedance-Based Modeling and Stability Analysis Approach for Power Oscillations Between Grid-Forming Inverters

This paper presents a new modeling and analysis approach to address the small-signal stability issues among interconnected grid-forming (GFM) inverters. A novel phasor domain “power impedance” model is developed to capture the terminal characteristics of GFM inverters. Based on the developed models, a sum-type impedance stability criterion is proposed to analyze the power oscillations between GFM Inverters. The developed models are validated with EMT simulations and the stability analysis results using the proposed stability criterion can match both EMT simulations and eigenvalue analysis. In addition, the application of the developed model can be extended to the stability analysis of large-scale power system with high penetration of GFM inverters.

Liu, Hanchao↗
DOE OSTI2021

Deep Reinforcement Learning Enabled Physical-Model-Free Two-Timescale Voltage Control Method for Active Distribution Systems

Active distribution networks are being challenged by frequent and rapid voltage violations due to renewable energy integration. Conventional model-based voltage control methods rely on accurate parameters of the distribution networks, which are difficult to achieve in practice. This paper proposes a novel physical-model-free two-timescale voltage control framework for active distribution systems. To achieve fast control of PV inverters, the whole network is first partitioned into several subnetworks using voltage-reactive power sensitivity. Then, the scheduling of PV inverters in the multiple sub-networks is formulated as Markov games and solved by a multi-agent soft actor-critic (MASAC) algorithm, where each subnetwork is modeled as an intelligent agent. All agents are trained in a centralized manner to learn a coordinated strategy while being executed based on only local information for fast response. For the slower time-scale control, OLTCs and switched capacitors are coordinated by a single agent-based SAC algorithm using the global information with considering control behaviors of the inverters. Particularly, the two-level agents are trained concurrently with information exchange according to the reward signal calculated from the data-driven surrogate model. Comparative tests with different benchmark methods on IEEE 33-and 123-bus systems and 342-node low voltage distribution system demonstrate that the proposed method can effectively mitigate the fast voltage violations and achieve systematical coordination of different voltage regulation assets without the knowledge of accurate system model.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2017

Crack Opening Displacement Behavior in Ceramic Matrix Composites

Ceramic Matrix Composites (CMC) modeling and life prediction strongly depend on oxidation, and therefore require a thorough understanding of when matrix cracks occur, the extent of cracking for given conditions (time-temperature-environment-stress), and the interactions of matrix cracks with fibers and interfaces. In this work, the evolution of matrix cracks in a melt-infiltrated Silicon Carbide/Silicon Carbide (SiC/SiC) CMC under uniaxial tension was examined using scanning electron microscopy (SEM) combined with digital image correlation (DIC) and manual crack opening displacement (COD) measurements. Strain relaxation due to matrix cracking, the relationship between COD's and applied stress, and damage evolution at stresses below the proportional limit were assessed. Direct experimental observation of strain relaxation adjacent to regions of matrix cracking is presented and discussed. Additionally, crack openings were found to increase linearly with increasing applied stress, and no crack was found to pass fully through the gage cross-section. This observation is discussed in the context of the assumption of through-cracks for all loading conditions and fiber architectures in oxidation modeling. Finally, the combination of SEM with DIC is demonstrated throughout to be a powerful means for damage identification and quantification in CMC's at stresses well below the proportional limit.

acoustic emission↗
NASA NTRS2016

InSitu SEM Investigation of Microstructural Damage Evolution and Strain Relaxation in a Melt Infiltrated SiC/SiC Composite

With CMC components poised to complete flight certification in turbine engines on commercial aircraft within the near future, there are many efforts within the aerospace community to model the mechanical and environmental degradation of CMCs. Direct observations of damage evolution are needed to support these modeling efforts and provide quantitative measures of damage parameters used in the various models. This study was performed to characterize the damage evolution during tensile loading of a melt infiltrated (MI) silicon carbide reinforced silicon carbide (SiC/SiC) composite. A SiC/SiC tensile coupon was loaded to a maximum global stress of 30 ksi in a tensile fixture within an SEM while observations were made at 5 ksi increments. Both traditional image analysis and DIC (digital image correlation) were used to quantify damage evolution. With the DIC analysis, microscale damage was observed at the fiber-matrix interfaces at stresses as low as 5 ksi. First matrix cracking took place between 20 and 25 ksi, accompanied by an observable relaxation in strain near matrix cracks. Matrix crack opening measurements at the maximum load ranged from 200 nm to 1.5 m. Crack opening along the fiber-matrix interface was also characterized as a function of load and angular position relative to the loading axis. This characterization was funded by NASA GRC and was performed to support NASA GRC modeling of SiC/SiC environmental degradation

testing↗