3+1S陀螺的故障检测方法研究

在收集各类陀螺故障模式的基础上,重点讨论了卫星出现大角速度超出陀螺测量范围而出现饱和的情况,并结合卫星质量特性,给出了避免误判的对策。同时以3+1S陀螺为例,从工程的角度,给出了确定平衡方程阈值的设置过程,并针对平衡方程对陀螺漂移故障的诊断存在延迟的情况进行了改进。仿真结果表明,通过设置两个阈值,使平衡方程不仅能对突变故障进行及时有效的诊断,也大大缩短了对陀螺漂移这样的缓变故障的诊断时间。     

涡喷发动机涡轮密封舱小气路流场数值计算

运用 NUMECA的三维粘性流动数值计算软件 FINE/ TURBO对涡喷发动机涡轮密封舱小气路中的复杂流场进行研究 ,计算中采用 Jameson的中心差分格式结合 Baldwin- Lomax湍流模型求解雷诺平均 N- S方程 ,得到了三维复杂流场中气体参数分布的详细结构和规律 ,具有较高的精度 ,为小气路中的故障的诊断提供数值依据     

民用飞机燃油泵的点火源防护

本文介绍民用飞机点火源防护的历史来源,提出接地故障保护的概念,分析接地故障中断继电器的基本工作原理及其在民用飞机上的应用。     

HW/SW Co-design of the Instrument Control Unit for the Energetic Particle Detector on-board Solar Orbiter

ESA’s medium-class Solar Orbiter mission is conceived to perform a close-up study of our Sun and its inner heliosphere to better understand the behaviour of our star. The mission will provide the clues to discover how the Sun creates and controls the solar wind and thereby affects the environments of all the planets. The spacecraft is equipped with a comprehensive suite of instruments. The Energetic Particle Detector (EPD) is one of the in-situ instruments on-board Solar Orbiter. EPD is composed of five different sensors, all of them sharing the Instrument Control Unit or ICU that is the sole interface with the spacecraft. This paper emphasises on how the hardware/software co-design approach can lead to a decrease in software complexity and highlights the versatility of the toolset that supports the development process. Following a model-driven engineering approach, these tools are capable of generating the high-level code of the software application, as well as of facilitating its configuration control and its deployment on the hardware platforms used in the different stages of the development process. Moreover, the use of the Leon2ViP virtual platform, with fault injection capabilities, allows an early software-before-hardware verification and validation and also a hardware–software co-simulation. The adopted solutions reduce development time without compromising the whole process reliability that is essential to the EPD success.     

一类飞行器网络控制系统的鲁棒故障检测算法

针对同时存在马尔可夫短时延和数据包丢失的网络环境,研究了一类飞行器网络控制系统的鲁棒故障检测问题。通过将马尔可夫短时延与丢包建模为一个有限马尔可夫链,并针对丢包建立相应的数学模型,飞行器网络控制系统被建模为离散有限马尔可夫跳变线性系统。在此基础上构建残差发生器,相应的故障检测问题归结为滤波问题。以线性矩阵不等式的形式给出并证明了故障检测滤波器的存在条件和求解方法。仿真结果表明,所提出的建模方法能够有效地减少丢包对故障检测性能的影响,同时上述故障检测方法对故障敏感,对未知扰动具有鲁棒性。     

A physics-based framework for online surface roughness assessment for high-pressure turbines

Surface roughness is a critical health parameter of a turbine blade due to its implications on blade surface heat transfer and structural integrity. This paper proposes a physics-based online framework for Gas Turbine Engines (GTE), in order to assess the blade surface roughness in a high-pressure turbine without engine shutdown. The framework consolidates Gas Path Analysis (GPA) based performance monitoring models and meanline turbomachinery analysis, using a novel GPA-meanline matching process. This extracts meaningful performance deviation trends from GPA, while addressing the uncertainties associated with the measurements and modelling. To relate efficiency loss to surface roughness severity, a meanline-based system-identification process has been developed to establish the meanline representation of the turbine stage, and to incorporate the empirical surface roughness loss correlations. The roughness loss correlations have been evaluated against recent transonic test data in the literature. A modification to the compressibility correction factor has been made according to the evaluation outcome, which improved loss predictions compared to the experimental measurements. The framework was tested on the three-year operational data of a cogeneration GTE, and the results verified the framework's potential for online surface roughness monitoring. The predicted surface roughness showed agreement in both trend and the magnitude-level with the measurements reported in the literature.     

一类非线性系统的故障可诊断性量化评价方法

为将提高卫星控制系统故障诊断能力的工作重点前移到地面设计阶段,针对其中的关键问题--可诊断性评价,提出一种适用于非线性模型的可诊断性量化评价方法,对传统定性评价研究进行了有效的扩展和延伸。鉴于线性化处理导致非线性因素引起的故障被忽略,将卫星控制系统描述成一类仿射非线性模型;并基于微分几何理论,给出可诊断性（包括可检测性和可隔离性）的定义和定性评价指标--不变最小对偶分布。通过子空间相似度判别准则,根据故障方向矢量和所得不变分布以及不同故障方向矢量之间的夹角关系,分别设计可检测性和可隔离性的量化指标,并给出具体评价流程;该指标能够明确故障被检测和隔离的难易程度,并可用于指导诊断算法的设计和传感器的优化配置。最后,以动量轮为仿真实例校验本文所提方法的正确性;仿真结果表明：该方法在不依赖任何诊断算法和具体故障模式的前提下,可以实现可检测性和可隔离性的准确量化评价。     

基于可观测性和可靠性的传感器分布优化设计

为了解决航天器的故障检测的传感器分布的设计问题,本文提出一种描述系统故障的有向图模型,针对该模型,给出了基于可观测性和可靠性的传感器分布的设计问题的描述,并提出了优化设计的思想。根据该思想给出了传感器分布优化设计方案,该方案考虑了故障的可观测性和故障检测的可靠性问题,还有传感器安置成本的约束问题,并采用了贪婪启发式算法实现了该方案。该方案应用到某卫星一次电源系统,仿真算例的结果表明该方案满足了可观测性和可靠性的要求,通过与其它设计方案对比,该方案能够快速提高系统的可靠性,更适合系统设计的需要。
     

卫星姿态控制系统故障重构观测器设计

对于卫星姿态控制系统,提出一种基于PD型学习观测器(Learning observer, LO)的系统故障重构方法。在P型LO的学习算法基础上引入测量输出估计误差的微分项,设计了一种PD型LO,估计卫星姿态角速度和姿态角的同时,快速精确重构卫星执行机构故障。给出了所提观测器的稳定性条件,并基于线性矩阵不等式技术提出一种系统化PD型LO设计方法。进一步,将所提PD型LO设计扩展用于卫星姿态敏感器故障的快速重构。最后,将所提方法应用于微小卫星推力器故障重构和陀螺故障重构,仿真结果校验了所提方法的有效性。     

基于故障树分析法的APU故障诊断研究

故障树分析（FTA）是故障分析和故障诊断中广泛应用的一种方法。结合APU的工作特性,对APU启动困难的原因进行了分析,建立APU启动困难故障树,并对故障树的基本原因事件进行了定性、定量分析,进而提出了排除APU故障的方法,以达到预测与预防故障发生的目的。