Development of a high pressure-differential turbine blade cooling system

A convective-film system of high pressure-differential turbine blade cooling is presented. The results of calculating the thermal-hydraulic blade state using the KW3D software are given.     

基于质量损失函数的模块实例稳健选配方法

为提高飞机工艺装备模块化设计的质量稳健性,提出了一种基于质量损失函数的模块化产品模块实例的稳健选配方法。该方法首先在给出质量损失函数统一描述的基础上,引入信噪比表达模块实例的稳定性。并综合考虑模块的功构连接形式和装配偏差,建立面向模块化产品的模块实例稳健选配模型,同时给出模型求解方法。最后,通过实例应用验证了方法的有效性和合理性。     

基于属性标记的关系数据库水印方法

数据库水印是保护数据库版权的一项重要技术。针对现有数据库水印方法无法抵抗有序关系模式变换攻击(ORSM)的问题,提出了一种基于属性标记的关系数据库水印方法。首先给出了有序关系模式变换攻击的定义,然后通过分析现有关系数据库水印算法无法抵抗ORSM攻击的问题,提出一种能够抵抗ORSM攻击的水印算法ALBWM,包括水印嵌入算法(ALBWM-EM)和水印检测算法(ALBWM-DT)。实验结果表明,与传统数据库水印算法相比,ALBWM算法在不影响性能的前提下对ORSM攻击具有更好的鲁棒性。     

数据挖掘在运载火箭智能测试中的应用

针对运载火箭复杂系统的故障检测难以建立准确的数学模型的问题,研究了基于数据驱动的数据挖掘异常检测算法,对多种数据挖掘算法在运载火箭发动机异常检测的应用进行了研究和分析,提出了基于混合概率密度统计的多策略异常检测评价算法。该算法基于非监督学习的算法挖掘火箭发动机不同参数间的正常关联模型,火箭发动机早期的异常数据会引起正常关联模型的破坏,引入混合概率密度统计的多策略异常检测评价机制,可以有效屏蔽参数测量故障对系统故障检测的影响,从而更加准确给出系统异常程度。使用发动机历史试车数据作为样本进行特征模型的训练,使用一元、多元和混合概率密度模型对存在异常的发动机试车数据进行了实时异常检测的实验验证。实验结果表明,相比传统基于阈值和规则的异常检测算法,基于概率密度统计的多策略异常检测算法不仅可给出系统的正常和异常的状态,还可计算各参数和整个系统的异常值,为运载火箭进一步的故障诊断提供更加灵活的参考。     

用于薄片试样弹塑性应力-应变分析的半解析方法

针对圆弧漏斗薄片试样和圆环薄片试样提出了半解析统一模型,并进行了有限元验证,该模型对漏斗薄片试样与圆环薄片试样单调力学行为均有优良的描述性。完成了漏斗薄片试样与圆环薄片试样的单轴拉伸试验及变幅对称循环试验,获得了载荷-位移曲线和循环稳定载荷幅-位移幅曲线。通过该模型并利用试验曲线预测材料单轴应力-应变关系和循环应力-应变关系,发现根据两种薄片试样预测的材料应力-应变关系及循环应力-应变关系与标准圆棒试样结果吻合良好。     

正六面体工装误差对光纤惯组加速度计零偏标定精度影响分析

研究了利用大理石平板正六面体工装对光纤惯组进行标定时,工装误差对标定精度的影响。首先建立了标定模型,设计了12位置法的标定方案与解算方法,其次分析了工装误差对标定影响的量化关系,并提出减小误差的标定解算方法,最后对理论分析结果进行仿真验证。仿真结果表明,当长宽高为400mm的工装平面度、垂直度、平行度误差均为0.04时,加速度计零偏的标定误差达到0.0001g,与工装误差角大小量级相当。用12位置二向法解算零偏可减小该误差影响。     

基于星敏感器的卫星姿态估计方法研究

以星敏感器姿态信息作为测量数据,结合卫星姿态动力学方程组,采用增广卡尔曼滤波方法对无陀螺卫星姿态进行估计。研究发现,干扰力矩的变化会降低卫星姿态估计结果的精度,为此提出两种改进的姿态估计算法:增广自适应卡尔曼滤波方法和增广强跟踪卡尔曼滤波方法;仿真表明,两种算法都能很好地克服干扰力矩变化导致的精度下降现象。     

Single Degenerate Models for Type Ia Supernovae: Progenitor’s Evolution and Nucleosynthesis Yields

We review how the single degenerate models for Type Ia supernovae (SNe Ia) works. In the binary star system of a white dwarf (WD) and its non-degenerate companion star, the WD accretes either hydrogen-rich matter or helium and undergoes hydrogen and helium shell-burning. We summarize how the stability and non-linear behavior of such shell-burning depend on the accretion rate and the WD mass and how the WD blows strong wind. We identify the following evolutionary routes for the accreting WD to trigger a thermonuclear explosion. Typically, the accretion rate is quite high in the early stage and gradually decreases as a result of mass transfer. With decreasing rate, the WD evolves as follows: (1) At a rapid accretion phase, the WD increase its mass by stable H burning and blows a strong wind to keep its moderate radius. The wind is strong enough to strip a part of the companion star’s envelope to control the accretion rate and forms circumstellar matter (CSM). If the WD explodes within CSM, it is observed as an “SN Ia-CSM”. (X-rays emitted by the WD are absorbed by CSM.) (2) If the WD continues to accrete at a lower rate, the wind stops and an SN Ia is triggered under steady-stable H shell-burning, which is observed as a super-soft X-ray source: “SN Ia-SSXS”. (3) If the accretion continues at a still lower rate, H shell-burning becomes unstable and many flashes recur. The WD undergoes recurrent nova (RN) whose mass ejection is smaller than the accreted matter. Then the WD evolves to an “SN Ia-RN”. (4) If the companion is a He star (or a He WD), the accretion of He can trigger He and C double detonations at the sub-Chandrasekhar mass or the WD grows to the Chandrasekhar mass while producing a He-wind: “SN Ia-He CSM”. (5) If the accreting WD rotates quite rapidly, the WD mass can exceed the Chandrasekhar mass of the spherical WD, which delays the trigger of an SN Ia. After angular momentum is lost from the WD, the (super-Chandra) WD contracts to become a delayed SN Ia. The companion star has become a He WD and CSM has disappeared: “SN Ia-He WD”. We update nucleosynthesis yields of the carbon deflagration model W7, delayed detonation model WDD2, and the sub-Chandrasekhar mass model to provide some constraints on the yields (such as Mn) from the comparison with the observations. We note the important metallicity effects on ⁵⁸Ni and ⁵⁵Mn.     

Crustal S-Wave Velocity from Apparent Incidence Angles: A Case Study in Preparation for InSight

Retrieval of crustal structure and thickness of Mars is among the main goals of InSight. Here we investigate which constraints on the crust at the landing site can be provided by apparent P-wave incidence angles derived from P-receiver functions. We consider receiver functions for six different Mars models, calculated from synthetic seismograms generated via Instaseis from the Green’s function databases of the Marsquake Service, in detail. To allow for a larger range of crustal thicknesses and structures, we additionally analyze data from five broad-band stations across Central Europe. We find that the likely usable epicentral distance range for P-wave receiver functions on Mars lies between \(35^{\circ}\) and the core shadow, and can be extended to more than \(150^{\circ}\) by also using the PP-phase. Comparison to models for the spatial distribution of Martian seismicity indicates that sufficient seismicity should occur within the P-wave distance range around InSight within the nominal mission duration to allow for the application of our method. Apparent P-wave incidence angles are derived from the amplitudes of vertical and radial receiver functions at the P-wave onset within a range of period bands, up to 120 s. The apparent incidence angles are directly related to apparent S-wave velocities, which are inverted for the subsurface S-wave velocity structure via a grid search. The veracity of the forward calculated receiver functions and apparent S-wave velocities is ensured by benchmarking various algorithms against the Instaseis synthetics. Results indicate that apparent S-wave velocity curves provide valuable constraints on crustal thickness and structure, even without any additional constraints, and considering the location uncertainty and limited data quantity of InSight. S-wave velocities in the upper half of the crust are constrained best, but if reliable measurements at long periods are available, the curves also provide constraints down to the uppermost mantle. Besides, it is demonstrated that the apparent velocity curves can differentiate between crustal velocity models that are indistinguishable by other methods.