人工神经网络在复合材料研究中的应用

人工神经网络技术广泛应用于复杂系统的建模中,已成为材料科学研究中常用的建模方法;介绍了BP神经网络及其建模的重要特征,综述了神经网络在复合材料性能预测、工艺设计与优化、损伤预测与检测等方面的研究情况.     

钨铼热电偶特性与试验

论述了钨铼热电偶的特性及测试方法。对钨铼热电偶采用几种不同的测试方法进行实际测试,得出的结果基本一致。同时,对影响测试结果的某些因素进行了分析。     

Analysis and modeling of information flow and distributed expertise in space-related operations

Evolving space operations requirements and mission planning for long-duration expeditions require detailed examinations and evaluations of information flow dynamics, knowledge-sharing processes, and information technology use in distributed expert networks. This paper describes the work conducted with flight controllers in the Mission Control Center (MCC) of NASA's Johnson Space Center. This MCC work describes the behavior of experts in a distributed supervisory coordination framework, which extends supervisory control/command and control models of human task performance. Findings from this work are helping to develop analysis techniques, information architectures, and system simulation capabilities for knowledge sharing in an expert community. These findings are being applied to improve knowledge-sharing processes applied to a research program in advanced life support for long-duration space flight. Additional simulation work is being developed to create interoperating modules of information flow and novice/expert behavior patterns.     

基于故障树模型的航天器故障诊断研究

本文建立了基于故障树模型的航天器故障诊断框架，提出了航天器故障诊断的知识组织和表示方法及相应的推理控制策略，并通过对某卫星能源系统故障实验台的故障诊断验证了方法的正确性。     

美国Dexter车轴公司零件项目国产材料的选用

对美国Dexter车轴公司芯轴零件材料及其在生产过程中暴露的问题进行了分析，提出了选用国产LHS-1材料进行零件生产的建议。通过对LHS-1材料及零件的性能测试和大批量生产的考核证明，用国产LHS-1低合金高强度结构钢生产的零件完全能够满足美国Dexter车轴公司芯轴零件的使用要求，在生产工艺上省去了调质热处理工序，显著降低了生产成本，获得了很好的经济效益。     

颗粒增强铝基复合材料研究与应用发展

简要分析了颗粒增强铝基复合材料的性能优势,综述了国外颗粒增强铝基复合材料的制备技术、性能水平、工程化应用技术以及应用发展现状,总结了国内工程化应用研究现状,提出了铝基复合材料在航空航天领域的应用发展方向.     

可用于平流层飞艇蒙皮的PBO织物编织和性能研究

文章根据平流层飞艇蒙皮的特性,研究聚对苯撑苯并二恶唑(PBO)纤维织物的编织方式.自制了平纹、斜纹和两经两纬织法的高强、轻质PBO平流层飞艇蒙皮织物,并进行厚度、面密度、拉伸强度和撕裂强度测试.研究表明,平纹织法更适合于编织PBO蒙皮织物,其工艺简单,编织的织物品质稳定,性能优异,面密度仅为86.45g/m2,拉伸强度大于1300N/cm,撕裂强度大于419N,可以满足平流层飞艇蒙皮材料的质量和强度要求.     

Continual learning fault diagnosis:A dual-branch adaptive aggregation residual network for fault diagnosis with machine increments

As a data-driven approach, Deep Learning(DL)-based fault diagnosis methods need to collect the relatively comprehensive data on machine fault types to achieve satisfactory performance. A mechanical system may include multiple submachines in the real-world. During condition monitoring of a mechanical system, fault data are distributed in a continuous flow of constantly generated information and new faults will inevitably occur in unconsidered submachines, which are also called machine increments....