Piloting a network of small telescopes

The Pilot Small Telescope Network (PSTN) is a state‐of the art system of easily replicable and scalable hardware, software, servers, eXtensible Markup Language (XML) protocols, and network middleware connecting and developing a pilot array of robotic telescopes to one another and the user community. The PSTN is a developmental project that will allow growing access to these telescopes, and make available data to faculty, students and others in an environment of collaboration. The underlying goal of the PSTN is to broaden the quantity and quality of astronomical education and research, particularly with a focus on traditionally underserved populations. (© 2006 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

近海污损生物的调查方法

近海海区污损生物调查有浮标挂板法和海上设施（浮标及其锚碇系统，Ｍａｒｅｘ水文浮标，移动式钻井平台和固定式生产平台）采样法。由于近海海区污损生物调查工作的特殊性和艰难性，笔者提出了利用已有的海上设施，尤其是固定式平台来积累近海污损生物资料，以便进行生态学研究。     

SELFE: A semi-implicit Eulerian–Lagrangian finite-element model for cross-scale ocean circulation

Unstructured-grid models grounded on semi-implicit, finite-volume, Eulerian–Lagrangian algorithms, such as UnTRIM and ELCIRC, have enjoyed considerable success recently in simulating 3D estuarine and coastal circulation. However, opportunities for improving the accuracy of this type of models were identified during extensive simulations of a tightly coupled estuary–plume–shelf system in the Columbia River system. Efforts to improve numerical accuracy resulted in SELFE, a new finite-element model for cross-scale ocean modeling. SELFE retains key benefits, including computational efficiency of existing semi-implicit Eulerian–Lagrangian finite-volume models, but relaxes restrictions on grids, uses higher-order shape functions for elevation, and enables superior flexibility in representing the bathymetry. Better representation of the bathymetry is enabled by a novel, “localized” vertical grid that resembles unstructured grids. At a particular horizontal location, SELFE uses either S coordinates or SZ coordinates, but the equations are consistently solved in Z space. SELFE also performs well relative to volume conservation and spurious oscillations, two problems that plague some finite-element models. This paper introduces SELFE as an open-source code available for community use and enhancement. The main focus here is on describing the formulation of the model and on showing results for a range of progressively demanding benchmark tests. While leaving details to separate publications, we also briefly illustrate the superior performance of SELFE over ELCIRC in a field application to the Columbia River estuary and plume.     

Study of Star/Galaxy Classification Based on the XGBoost Algorithm

Machine learning has achieved great success in many areas today. The lifting algorithm has a strong ability to adapt to various scenarios with a high accuracy, and has played a great role in many fields. But in astronomy, the application of lifting algorithms is still rare. In response to the low classification accuracy of the dark star/galaxy source set in the Sloan Digital Sky Survey (SDSS), a new research result of machine learning, eXtreme Gradient Boosting (XGBoost), has been introduced. The complete photometric data set is obtained from the SDSS-DR7, and divided into a bright source set and a dark source set according to the star magnitude. Firstly, the ten-fold cross-validation method is used for the bright source set and the dark source set respectively, and the XGBoost algorithm is used to establish the star/galaxy classification model. Then, the grid search and other methods are used to adjust the XGBoost parameters. Finally, based on the galaxy classification accuracy and other indicators, the classification results are analyzed, by comparing with the models of function tree (FT), Adaptive boosting (Adaboost), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), Stacked Denoising AutoEncoders (SDAE), and Deep Belief Nets (DBN). The experimental results show that, the XGBoost improves the classification accuracy of galaxies in the dark source classification by nearly 10% as compared to the function tree algorithm, and improves the classification accuracy of sources with the darkest magnitudes in the dark source set by nearly 5% as compared to the function tree algorithm. Compared with other traditional machine learning algorithms and deep neural networks, the XGBoost also has different degrees of improvement.     

A method for aligning the plastic scintillator detector on DAMPE

The Plastic Scintillator Detector(PSD) onboard the DArk Matter Particle Explorer(DAMPE)is designed to measure cosmic ray charge(Z) and to act as a veto detector for gamma ray identification.To fully exploit the charge identification potential of PSD and to enhance its capability to identify gamma ray events, we develop an alignment method for the PSD. The path length of a given track in the volume of a PSD bar is derived taking into account the shift and rotation alignment corrections. By examining energy spectra of corner-passing events and fully contained events, position shifts and rotations of all PSD bars are obtained, and are found to be on average about 1 mm and 0.0015 radian respectively. To validate the alignment method, we introduce artificial shifts and rotations of PSD bars into the detector simulation.These shift and rotation parameters can be recovered successfully by the alignment procedure. As a result of the PSD alignment procedure, the charge resolution of the PSD is improved from 4% to 8%, depending on the nuclei.     

基于EfficientNet的星系形态分类研究

星系的结构和形态能够反映星系自身的物理性质,其形态的分类是后续分析研究的一个重要环节.EfficientNet模型使用复合系数对深度网络模型的深度、宽度、输入图像分辨率进行更加结构化的统一缩放,是一种新的深度网络优化扩展方法.将该模型应用于星系数据形态的分类研究中,结果表明基于EfficientNetB5模型的平均准确率、精确率、召回率以及F1分数(精确率与召回率的调和平均数)都在96.6%以上,与残差网络(Residual network, ResNet)中ResNet-26模型的分类结果相比有较大的提升.实验结果证明EfficientNet的深度网络优化扩展方法可行且有效,可应用于星系的形态分类.     

考虑LS+AR模型基础数据量的极移预报研究

针对目前极移最小二乘(Least Square, LS)+自回归(AutoRegressive, AR)预报模型的单一数据选取方案, 提出分别考虑LS模型数据量和AR残差数据量的组合数据模式, 并对极移预报时单一数据和组合数据预报结果精度进行分析, 探讨模型输入数据量对极移预报精度的影响. 结果表明, 模型输入数据量的变化对极移预报结果影响较大. 采用组合数据预报的方式相比较于单一数据量预报方式精度更高, 特别是针对30--360 d跨度内的中长期预报, 组合数据量的极移预报精度可比单一数据量预报精度有较大改善. 结论证明组合数据在极移预报时具有一定的优势, 可为以后极移预报数据量选取提供一定的借鉴参考意义.     

Performance assessment of a concept propulsor: the thrust-balanced propeller

This paper presents the results of a numerical performance analysis to demonstrate the worthiness of a recently patented new concept propulsor, the so-called “thrust-balanced propeller (TBP)”. The main advantage of this unconventional propulsor is its inherent ability to reduce the unsteady effect of blade forces and moments when it is operating in a non-uniform wake flow. The propulsor comprises a pair of diametrically opposed blades that are connected to one another and mounted so as to be rotatable together through a limited angle about their spindle axis. A quasi-hydrodynamic approach is described and applied to perform the numerical analysis using a state-of-the-art lifting surface procedure for conventional propellers. Performance comparisons with a conventional fixed-pitch propeller are made for the blade forces and moments, efficiency, cavitation extents and fluctuating hull pressures. Bearing in mind the quasi-static nature of the analyses, the results present favourable performance characteristics for the thrust-balanced propeller and support the worthiness of the concept. However, the concept needs to be proved through physical model tests, which are planned to take in a cavitation tunnel.