Boosting the electron beam transmittance of field emission cathode using a self-charging gate.

The gate-type carbon nanotubes cathodes exhibit advantages in long-term stable emission owing to the uniformity of electrical field on the carbon nanotubes, but the gate inevitably reduces the transmittance of electron beam, posing challenges for system stabilities. In this work, we introduce electron beam focusing technique using the self-charging SiNx/Au/Si gate. The potential of SiNx is measured to be approximately -60 V quickly after the cathode turning on, the negative potential can be maintained as the emission goes on. The charged surface generates rebounding electrostatic forces on the following electrons, significantly focusing the electron beam on the center of gate hole and allowing them to pass through gate with minimal interceptions. An average transmittance of 96.17% is observed during 550 hours prototype test, the transmittance above 95% is recorded for the cathode current from 2.14 µA to 3.25 mA with the current density up to 17.54 mA cm-2.

TTN: Topological Transformer Network for Automated Coronary Artery Branch Labeling in Cardiac CT Angiography.

OBJECTIVE: Existing methods for automated coronary artery branch labeling in cardiac CT angiography face two limitations: 1) inability to model overall correlation of branches, since differences between branches cannot be captured directly. 2) a serious class imbalance between main and side branches. METHODS AND PROCEDURES: Inspired by the application of Transformer in sequence data, we propose a topological Transformer network (TTN), which solves the vessel branch labeling from a novel perspective of sequence labeling learning. TTN detects differences between branches by establishing their overall correlation. A topological encoding that represents the positions of vessel segments in the artery tree, is proposed to assist the model in classifying branches. Also, a segment-depth loss is introduced to solve the class imbalance between main and side branches. RESULTS: On a dataset with 325 CCTA, our method obtains the best overall result on all branches, the best result on side branches, and a competitive result on main branches. CONCLUSION: TTN solves two limitations in existing methods perfectly, thus achieving the best result in coronary artery branch labeling task. It is the first Transformer based vessel branch labeling method and is notably different from previous methods. CLINICAL IMPACT: This Pre-Clinical Research can be integrated into a computer-aided diagnosis system to generate cardiovascular disease diagnosis report, assisting clinicians in locating the atherosclerotic plaques.

A newly designed ignition method for miniature radio frequency ion thruster.

The ignition methods used by micro-radio frequency (RF) ion thrusters have a disadvantage, that is, the starting voltage and flow rate are obviously higher than the rated value, which will easily damage the grid after long-term use. To decrease the starting voltage and flow and reduce the damage to the grid, a new ignition system is proposed in this paper. This system uses an intake pipe as the ground electrode, has an inductance coil and a pre-ionization chamber, and enables the miniature RF ion thruster (Harbin Institute of Technology's RF Ion Thruster 4, HRIT-4) to ignite at the rated voltage and flow by means of strong electric field breakdown and electromagnetic coupling. The experimental results show that when the flow rate is 1.0 SCCM, the ignition voltage is lower than 1900 V, and when the flow rate is 1.5 SCCM, the ignition voltage is lower than 1400 V.

Throttling process of a supersonic cascade studied by high-frequency response pressure and high-speed schlieren.

In this study, a single-channel supersonic cascade model is investigated experimentally at a freestream Mach number of 2.4 to obtain a better understanding of the flow field evolution during the throttling process. A flap is placed at the channel exit to choke the flow linearly. Measurements include 1-kHz schlieren imaging and 10-kHz simultaneous fast-response wall pressure. Three stages, namely attached flow, separated flow, and oscillatory flow, are identified in the throttling process. The joint time-frequency analysis and wall pressure spectrum contour exhibit the time evolution and spatial distribution of the pressure fluctuation. With the increase in backpressure, the pressure fluctuation in the low-frequency shock oscillation range of 40-400 Hz on the suction surface located in the separated flow gradually enhances. The power spectral, coherence, and phase analyses of the schlieren images describe the dominant oscillation structure and its relationship with other regions. During the separated flow, the pressure change in the subsonic separated region first lead to a change in the state of the separated shear layer, after which the shock waves in the shock train, move. The oscillatory flow is a process wherein the upstream shock wave oscillates, causing the entire downstream channel to fluctuate.

Note: Possibilities of detecting the trace-level erosion products from an electric propulsion hollow cathode plasma source by the method of time-of-flight mass spectrometry.

A hollow cathode produces electrons which neutralize ions from electric propulsion thrusters. After hundreds to thousands of hours of operation in space, the cathode materials can be significantly eroded due to ion bombardment. As a result, the electric propulsion system performance will be obviously changed or even fail. In this work, the erosion products from a LaB6 hollow cathode (widely used presently in electric propulsion systems) are studied by using a specific detection system, which consists of a molecular beam sampler and a time-of-flight mass spectrometer. This system measures trace-level-concentration (10-6-10-3) products. Boron (B), tantalum (Ta), and tungsten (W)-originating from the emitter, keeper, and orifice of the hollow cathode-are measured. It is found that the erosion rate is significantly influenced by the gas flow rate to the cathode.

Stabilizing low-frequency oscillation with two-stage filter in Hall thrusters.

The use of a filter is the most common method to suppress low-frequency discharge current oscillation in Hall thrusters. The only form of filter in actual use involves RLC networks, which serve the purpose of reducing the level of conducted electromagnetic interference returning to the power processing unit, which is the function of a filter. Recently, the role of the filter in the oscillation control was introduced. It has been noted that the filter regulates the voltage across itself according to the variation of discharge current so as to decrease its fluctuation in the discharge circuit, which is the function of a controller. Therefore, a kind of two-stage filter is proposed to fulfill these two purposes, filtering and controlling, and the detailed design methods are discussed and verified. A current oscillation attenuation ratio of 10 was achieved by different capacitance and inductance combinations of the filter stage, and the standard deviation of low-frequency oscillations decreased from 3 A-1 A by the control stage in our experiment.

Note: An approach to measurement of low frequency oscillation amplitude of discharge current of in-orbit Hall thruster.

This paper provides a method to measure the amplitude of low frequency oscillation under the on-track working condition, and realizes the sampling by means of adding the circuit design of sampling, low pass filtering by 3 dB at 48.2 kHz, detection and integrating in the filtering unit. The experimental results prove that the measuring device of merely 0.8 g can quantitatively reflect the amplitude of low frequency oscillation in Hall thruster and the maximum deviation of experiment data and theory data is 10% FS.

Selecting discrete and continuous features based on neighborhood decision error minimization.

Feature selection plays an important role in pattern recognition and machine learning. Feature evaluation and classification complexity estimation arise as key issues in the construction of selection algorithms. To estimate classification complexity in different feature subspaces, a novel feature evaluation measure, called the neighborhood decision error rate (NDER), is proposed, which is applicable to both categorical and numerical features. We first introduce a neighborhood rough-set model to divide the sample set into decision positive regions and decision boundary regions. Then, the samples that fall within decision boundary regions are further grouped into recognizable and misclassified subsets based on class probabilities that occur in neighborhoods. The percentage of misclassified samples is viewed as the estimate of classification complexity of the corresponding feature subspaces. We present a forward greedy strategy for searching the feature subset, which minimizes the NDER and, correspondingly, minimizes the classification complexity of the selected feature subset. Both theoretical and experimental comparison with other feature selection algorithms shows that the proposed algorithm is effective for discrete and continuous features, as well as their mixture.