BP神经网络在电子设备可靠性验证中的应用

讨论了借助MATLAB的BP神经网络解决电子设备在寿命分布未知和小样本试验数据时的可靠性评估方法, 并给出了一个应用实例, 该方法可以提高可靠性评估置信度, 并能节省试验时间和费用, 特别适用于大型复杂电子设备。     

车辆液力机械传动系统建模及动态特性仿真

建立了具有液力传动的履带车辆动力传动系统及整车的仿真模型,对车辆的加速性进行了仿真研究并与试验结果进行了比较,应用该仿真模型,分析了履带车辆换挡过程中发动机与涡轮转速、系统元件转速等的变化规律,结果表明该仿真模型能够方便的应用于车辆的性能分析与预测.     

基于改进的混合高斯模型的背景建模方法

提出了一种基于改进的混合高斯模型的背景建模方法,克服了经典混合高斯模型方法计算量大和对长时间静止物体转为运动及光照突变较为敏感的缺点。首先,在经典混合高斯模型方法的基础上,引入了一种新的高斯分布个数的自适应选择策略,提高了建模效率。其次,分析了经典混合高斯模型方法对长时间静止物体转为运动及光照突变较为敏感的原因,采用了一种不同区域更新率的自适应选择策略,能够迅速响应场景的变化,有效地解决了大面积误检问题。通过在典型的场景下与经典混合高斯模型方法进行比较,验证了本文算法的有效性。     

应用AHP法分析地域通信网可靠性

研究了地域通信网的可靠性,提出应用AHP法简化地域通信网可靠性研究的复杂度,给出一种适于通信网可靠性求解的AHP算法,最后结合装甲兵地域通信网的应用,分析了一个地域通信网可靠性实例,说明各层因素对整个通信网可靠性的综合作用。     

热力设备可靠性分析的问题与对策

以热力设备为研究对象 ,从可靠性分析特点出发 ,对应用故障统计方法所遇到的典型问题作了探讨 ,就分析方法、建模技术等方面论述了作者的观点     

Life distribution analysis based on Lévy subordinators for degradation with random jumps

For a component or a system subject to stochastic degradation with sporadic jumps that occur at random times and have random sizes, we propose to model the cumulative degradation with random jumps using a single stochastic process based on the characteristics of Lévy subordinators, the class of nondecreasing Lévy processes. Based on the inverse Fourier transform, we derive a new closed‐form reliability function and probability density function for lifetime, represented by Lévy measures. The reliability function derived using the traditional convolution approach for common stochastic models such as gamma degradation process with random jumps, is revealed to be a special case of our general model. Numerical experiments are used to demonstrate that our model performs well for different applications, when compared with the traditional convolution method. More importantly, it is a general and useful tool for life distribution analysis of stochastic degradation with random jumps in multidimensional cases. © 2015 Wiley Periodicals, Inc. Naval Research Logistics 62: 483–492, 2015     

序降应力加速寿命试验(下)——优化设计篇

如何设计试验方案使试验评估结果更准确、试验代价更小,是序降应力加速寿命试验应用中面临的一个重要问题。以试验评估精度和试验代价共同需求为背景,建立一类新的基于加权系数的优化目标函数。为解决仿真优化设计过程中的计算量问题,在深入分析目标函数性质基础上,提出一种新的优化设计法——正交搜索法。与优化设计过程的全局搜索法不同,该方法能够逐次缩小目标函数值,快速逼近最优试验方案。最后通过仿真算例验证了新方法的正确性和高效性。     

利用分时刷新和位置约束的卫星载荷BRAM抗辐照设计方法

为解决星载资源严格受限条件下,静态随机存取存储器型现场可编程门阵列中块随机存储器(block random access memory, BRAM)的轻量级、高可靠抗辐照加固难题,提出了一种基于分时刷新和位置约束的卫星载荷BRAM抗辐照加固设计方法,通过监控算法执行时隙实现BRAM分时刷新,并增加位置约束有效降低三模冗余后两模设计同时发生异常的概率,以较少的资源消耗有效提升BRAM在轨抗辐照可靠性。重离子加速试验结果表明,采用分时刷新和位置约束加固方法后,卫星载荷单粒子功能中断截面下降约81.63%,在轨BRAM异常由3颗星2年发生3次减少为25颗星2年未发生,在轨抗辐照可靠性大幅提升。     

利用整帧翻转的SRAM型FPGA故障注入加速算法

不可恢复异常比特位(unrecoverable-sensitive bits, UR-SB)不能通过定时刷新修复,会造成卫星载荷在轨服务的长时间中断,是地面故障注入试验需要着重评估和改善的。但UR-SB占比极低,若采用传统逐位翻转故障注入方法,其测试耗时太长,效率极低。提出了一种基于整帧翻转的静态随机存取存储器型现场可编程门阵列的故障注入加速算法,其通过整帧翻转能够快速筛除不存在UR-SB的配置帧,并进一步用二分法对存在UR-SB的配置帧进行快速搜索,有效加速了UR-SB的精确定位过程。以在轨常用XQR2V3000器件为例,理论分析在较差情况下测试效率可提升207倍,信号生成载荷实测结果最高可提升949倍,理论分析和实测结果均验证了所提加速算法的有效性。     

Two novel critical shock models based on Markov renewal processes

In this paper, we investigate systems subject to random shocks that are classified into critical and noncritical categories, and develop two novel critical shock models. Classical extreme shock models and run shock models are special cases of our developed models. The system fails when the total number of critical shocks reaches a predetermined threshold, or when the system stays in an environment that induces critical shocks for a preset threshold time, corresponding to failure mechanisms of the developed two critical shock models respectively. Markov renewal processes are employed to capture the magnitude and interarrival time dependency of environment-induced shocks. Explicit formulas for systems under the two critical shock models are derived, including the reliability function, the mean time to failure and so on. Furthermore, the two critical shock models are extended to the random threshold case and the integrated case where formulas of the reliability indexes of the systems are provided. Finally, a case study of a lithium-ion battery system is conducted to illustrate the proposed models and the obtained results.