Reliability analysis based on the principle of maximum entropy and Dempster–Shafer evidence theory

The Probability density functions (PDFs) of some uncertain parameters are difficult to determine precisely due to insufficient information. Only the varying intervals of such parameters can be obtained. A method of reliability analysis based on the principle of maximum entropy and evidence theory was proposed to address the reliability problems of random and interval parameters. First, the PDFs and cumulative distribution functions of interval parameters were obtained on the basis of the principle of maximum entropy and Dempster–Shafer evidence theory. Second, the normalized means and standard deviations of interval parameters were obtained using the equivalent normalization method. Third, two explicit iteration algorithms of reliability analysis were proposed on the basis of the advanced firstorder and second-moment method to avoid solving the limit state function and obtain the reliability index. Finally, the accuracy and efficiency of the proposed methods were verified through a numerical example and an engineering case.     

Numerical simulation of bubble separation length in a gas–liquid separator based on the Euler–Lagrange method

In this study, the bubble separation behavior in a gas–liquid separator is numerically investigated on the basis of the Euler–Lagrange approach, in which the forces acting on bubbles in a swirling flow field are modeled to calculate the trajectories of the bubbles. By adopting this approach, the effects of five parameters, namely, back pressure, Reynolds number, bubble diameter, void fraction, and swirl number, on separation performance in terms of pressure loss, separation efficiency, separation length, and split ratio are computed and analyzed. On the basis of the analysis, correlations of separation length with the two main parameters are established, which can serve as a basis for the optimal design of separator.     

Development of a radiative transfer model for the determination of toxic gases by Fourier transform–infrared spectroscopy with a support vector machine algorithm

Abstract

This report describes a radiative transfer model for Fourier transform-infrared (FT-IR) spectroscopy to create close-to-reality toxic gas spectra by reflecting the unique spectral responses of detectors and using the atmospheric radiative transfer code, MODTRAN. This system can be highly useful in overcoming the limitations for measuring toxic gases in open environments. The emulated gas spectra can be used to train support vector machine (SVM) for chemical gas detection. Its detection performance is evaluated with nerve agents (tabun, sarin, soman, and cyclosarin) and a simulant gas (sulfur hexafluoride) for indoor and outdoor experiments by using two off-the-shelf FT-IR gas detectors. The experimental results show that the proposed SVM algorithm successfully detected and classified targeted gases while reducing false negative and false positive detection rates.     

An experimental study on real–time analysis of two–phase peristaltic slug flows in dialysis machines

Two–phase flows appear in many industrial and biomedical applications. One of the most vital biomedical applications of two–phase flows is in hemodialysis machines due to air embolism and heparin injection. Since these flows have a very complex and intermittent nature, studying their dynamics is a very challenging and fundamental problem. The purpose of this article is to present an experimental study on the dynamics of two–phase peristaltic slug flows. The measurement strategy is based on the image processing technology. The characteristic parameters of the two–phase pulsatile slug flows, including the slug length, as well as the translational velocity and frequency of the slug motion, are measured, and the effect of the liquid flow rate and liquid superficial velocity is investigated. The results show that the average and maximum slug velocities, and also the dominant amplitude of the slug velocity increase with the flow rate and liquid superficial velocity, while it is not possible to clearly predict a correlation between the liquid superficial velocity and the slug length. The measurement strategy presented in this article can be used in the control and alarm systems of smart dialysis machines.     

Modeling and analysis of the effects of processing parameters on the performance characteristics in the high pressure die casting process of Al–SI alloys

The high pressure die casting (HPDC) process has achieved remarkable success in the manufacture of aluminum–silicon (Al–SI) alloy components for the modern metal industry. Mathematical models are proposed for the modeling and analysis of the effects of machining parameters on the performance characteristics in the HPDC process of Al–SI alloys which are developed using the response surface methodology (RSM) to explain the influences of three processing parameters (die temperature, injection pressure and cooling time) on the performance characteristics of the mean particle size (MPS) of primary silicon and material hardness (HBN) value. The experiment plan adopts the centered central composite design (CCD). The separable influence of individual machining parameters and the interaction between these parameters are also investigated by using analysis of variance (ANOVA). With the experimental values up to a 95% confidence interval, it is fairly well for the experimental results to present the mathematical models of both the mean particle size of primary silicon and its hardness value. Two main significant factors involved in the mean particle size of primary silicon are the die temperature and the cooling time. The injection pressure and die temperature also have statistically significant effect on microstructure and hardness.     

Design and analysis of a new AUV’s sliding control system based on dynamic boundary layer

The new AUV driven by multi-vectored thrusters not only has unique kinematic characteristics during the actual cruise but also exists uncertain factors such as hydrodynamic coefficients perturbation and unknown interference of tail fluid,which bring difficult to the stability of the AUV’s control system.In order to solve the nonlinear term and unmodeled dynamics existing in the new AUV’s attitude control and the disturbances caused by the external marine environment,a second-order sliding mode controller with double-loop structure that considering the dynamic characteristics of the rudder actuators is designed,which improves the robustness of the system and avoids the control failure caused by the problem that the design theory of the sliding mode controller does not match with the actual application conditions.In order to avoid the loss of the sliding mode caused by the amplitude and rate constraints of the rudder actuator in the new AUV’s attitude control,the dynamic boundary layer method is used to adjust the sliding boundary layer thickness so as to obtain the best anti-chattering effects.Then the impacts of system parameters,rudder actuator’s constraints and boundary layer on the sliding mode controller are computed and analyzed to verify the effectiveness and robustness of the sliding mode controller based on dynamic boundary layer.The computational results show that the original divergent second-order sliding mode controller can still effectively implement the AUV’s attitude control through dynamically adjusting the sliding boundary layer thickness.The dynamic boundary layer method ensures the stability of the system and does not exceed the amplitude constraint of the rudder actuator,which provides a theoretical guidance and technical support for the control system design of the new AUV in real complex sea conditions.     

The influence of a change in the electron trajectory in the vacuum-diode anode–cathode gap on the impedance

The analysis of the influence of the electron-trajectory curvature in the anode–cathode gap of a vacuum diode on its impedance during generation of high-current electron beams is performed, and the applicability of the one-dimensional Child–Langmuir formula for the electron-current calculation is evaluated. The results of an experimental study of a flat diode with explosive-emission graphite, copper, and carbon- fabric cathodes and also with a multipointed cathode are presented. These investigations were performed on the TEU-500 accelerator (350–500 kV, 60 ns). A strip diode with a graphite cathode was also studied in the mode of magnetic self-insulation of electrons. The experiments were performed on the TEMP-4М accelerators (150–200 kV, 400–600 ns). The investigation results showed that the satisfactory coincidence of the experimental values of the total current with those calculated from the one-dimensional Child–Langmuir formula (for the working area of the cathode) is observed in diodes not only in the absence of a change in the electron trajectory in the anode–cathode gap but also upon a deflection of the trajectory from the normal to the cathode surface by an angle of <90°. An increase in the electron current owing to a decrease in the anode–cathode gap and an increase in the emission area of the cathode during the cathode-plasma production and also at the expense of the presence of microirregularities on the cathode is properly described by the onedimensional Child-Langmuir formula, if the above factors are taken into account.     

Effects of strain induced martensitic transformation on the cavitation erosion resistance and incubation time of Fe–Cr–Ni–C alloys

The effect of a strain-induced martensitic transformation on the cavitation erosion resistance and incubation time of Fe–10Cr–10Ni–xC (x = 0.3, 0.4, 0.5, and 0.6 wt%) austenitic steels has been studied. As the carbon concentration increased, mass loss in the alloys also increased, while the incubation period and the amount of transformed martensite decreased. In addition, the martensite volume fraction increased with increasing testing time and reached a saturation point for each test alloy. After the saturation point was reached, the martensite volume fraction did not change throughout the remainder of the test, even though the transformed martensite phase was removed. This result indicates that new martensite phases were formed immediately after the removal of the previously formed martensite. Martensitic transformation exerts significant effects on wear resistance and incubation time by steadily absorbing the cavity collapse energy.     

Measurement of electromagnetic radiation with respect to the hours and days of a week at 100kHz–3GHz frequency band in a turkish dwelling

In this paper, the electromagnetic (EM) radiation to which we are exposed in our life is aimed to be observed during a week. The measurements are carried out in a randomly selected apartment during 24 h a day at 4 s sampling period for all days of a week. Due to the result of the study, we can see whether any radiation above specified limits occurs during a day and week or not and also we can determine how an ordinary instantaneous measurement represents the usual radiation level in related area. Measurement results showed that EM radiation levels in 100 kHz–3 GHz frequency band were considerably below specified limit values. There is statistically no difference between measured values in Monday and Saturday with respect to days. However, it is observed that the measured values in other days except for Monday and Saturday are statistically considerably different.     

Experimental method of fabrication of a matched metal–dielectric structure for a sensor based on the effect of frustrated total internal reflection

An experimental method of fabrication of a sensor based on a metal–dielectric structure (Al + ZnS) and optimization of its characteristics is described. The coefficient of light reflection (p-polarization) from the aluminum layer is studied as a function of the layer thickness for different angles of incidence at the wavelength of 532 nm. Based on calculations, which are qualitatively consistent with experimental results, a structure consisting of matched layers of aluminum and zinc sulfide is fabricated; this structure has a higher angular resolution than the aluminum film with no dielectric coating. The detection limit of angular measurements by the sensor based on this structure is estimated as 2.6 · 10^-5 RIU (refraction index units).     

基于小波分析和支持向量机的旋转机械故障诊断方法

提出了一种基于小波分析和支持向量机相结合的旋转机械故障诊断方法.首先运用小波包对振动信号进行分解和重构,然后提取各个频带里的信号能量值,将该能量值作为特征参数输入到支持向量机,进行故障模式识别.通过对实验数据的分析表明,与BP神经网络相比,该方法可以获得更高的旋转机械故障诊断准确率.     

旋转机械智能维护单元设计研究

分析了大型旋转机械在线监测与诊断系统应用中存在的标准化、模块化、集成化程度低 ,可靠性差以及综合数据利用率低等问题及其原因 ,为了满足对该系统的高可靠性、网络化和智能化要求 ,提出了多参数集成智能维护单元解决方案。提出了智能维护单元体结构模型 ,设计了基于智能维护单元的旋转机械监视诊断与维护系统 ,在单元体集成了传感器二次仪表、供电及接口电路模块 ;系统可在嵌入式处理器和DSP处理器的控制下完成设备数据采集、处理、特征提取、状态识别、故障诊断、数据管理和远程维修服务等功能。多参数智能维护单元体技术将成为旋转机械监测诊断系统采用的模式之一     

基于小波变换和ICA特征提取的开关电流电路故障诊断

提出了采用小波变换和独立成分分析(ICA)作为预处理器来进行特征提取的神经网络开关电流电路故障诊断方法。该方法对采集到的故障响应信号进行Haar小波正交滤波器分解,获得低频近似信息和高频细节信息;然后利用独立成分分析方法进行ICA故障特征提取;最后将所得到的最优故障特征输入到BP神经网络中进行故障分类。对六阶切比雪夫低通滤波器和六阶椭圆带通滤波器电路进行了仿真实验验证,获得了100%的故障诊断准确率,与其他方法进行比较,实验结果显示了该方法的优越性。     

基于矢谱和粗糙集理论的旋转机械故障诊断

矢谱融合了转子同源双通道的信息,能准确反映转子运动状态．粗糙集理论是一种对决策表进行简化,去除冗余属性的数据分析和处理方法．提出了基于矢谱和粗糙集理论的旋转机械故障诊断方法．计算了旋转机械振动4种典型故障的矢谱征兆,使用粗糙集理论对其进行约简,根据约简的结果生成矢谱诊断规则,并利用得到的规则对故障测试样本进行了诊断．结果表明：相对于单通道数据,基于矢谱和粗糙集理论的故障诊断不仅简化了诊断规则,而且明显提高了故障诊断的准确率．     

基于EMD的冲击信号提取方法及其在设备故障诊断中的应用

提出一种基于经验模式分解(Empirical Mode Decomposition,EMD)的冲击信号提取方法,利用该方法首先将含有周期性冲击的信号进行EMD分解,在分解后的高频段IMF中,存在着类似冲击响应信号的成分,这些成分是由原始信号中的周期性冲击引起的,通过包络解调方法,可以得到冲击响应信号出现的频率,该频率对应原信号中冲击信号出现的频率.由于碰摩故障发生时,往往伴随着周期性冲击信号的产生,故该方法可以应用于旋转设备碰摩故障诊断中.仿真信号和试验数据的分析结果表明,这种方法正确有效,可以应用于工程实际.     

基于Volterra级数的提升小波变换边界处理及应用

针对现有的提升小波变换容易产生边界振荡和频率混叠的不足,提出了一种将Volterra级数模型和抗混叠提升小波包相结合的信号处理方法.首先对信号两端进行数据延拓,用二阶Volterra级数预测模型对延拓信号进行预测;然后用抗混叠提升算法对信号进行小波包分解.对仿真信号进行边界处理后,信号在边界不会出现振荡现象;用抗混叠提升小波包对信号进行分解不会引起频率混叠现象.工程应用中,从强大的背景噪声中提取出了往复泵柱塞与缸套碰磨产生的微弱振动冲击信号,诊断出了密封盘根过度磨损的故障.     

基于DBN的故障特征提取及诊断方法研究

随着装备日趋复杂化,依靠专家经验或信号处理技术人工提取和选择故障特征变得越来越困难。此外,以BP神经网络、SVM为代表的浅层模型难以表征被测信号与装备健康状况之间复杂的映射关系,且面临维数灾难等问题。结合深度置信网络(DBN)在提取特征和处理高维、非线性数据等方面的优势,提出一种基于深度置信网络的故障特征提取及诊断方法。该方法通过深度学习利用原始时域信号训练深度置信网络并完成智能诊断,其优势在于能够摆脱对大量信号处理技术与诊断经验的依赖,完成故障特征的自适应提取与健康状况的智能诊断,该方法对时域信号没有周期性要求,具有较强的通用性和适应性。在仿真数据集和轴承数据集上进行了故障特征提取和诊断实验,实验结果表明:本文提出的方法能够有效地从原始信号中进行多种工况、多种故障位置和多种故障程度的故障特征提取和诊断,并且具有较高的故障识别精度。     

基于Internet的嵌入式设备状态监测系统开发与研究

网络化的设备状态监测及故障诊断是企业设备状态监测的发展方向。本文介绍了一种基于Internet的嵌入式设备状态监测与故障诊断系统，利用嵌入式动态信号测试分析仪进行现场数据采集，通过Internet上传至服务器，实现了设备数据的集中化管理，专家通过网络可以对设备状态进行远程监控，并且服务器可以对客户端的连接进行在线监控，增强了系统的规范化管理和安全性。     

脑电信号的混沌分析和小波包变换特征提取算法

针对脑电(EEG)信号的手部动作模式信息处理,提出一种混沌分析和小波包变换相结合的特征提取方法.用眼动辅助来采集手部动作时的脑电信号,对采集的C3、C4 、P3和P4脑电信号消噪后分别用混沌分析和小波包变换的方法进行特征提取,前者提取混沌特征的最大Lyapunov指数和关联维数,组成8维向量;后者提取脑电信号的4种特征节律波,分别计算其相对能量,组成16维向量;最后把两种方法提取的向量组成24维特征向量,输入SVM分类器,实现基于EEG信号的手部动作模式的识别.对不同个体上翻、下翻、展拳、握拳4种手部动作的识别实验表明,平均识别率均在80%以上,明显优于其他方法识别的结果.     

基于ARM的嵌入式系统在数控设备故障诊断中的应用

分析传统数控设备故障诊断系统所存在的问题，将ARM嵌入式系统引入传统数控设备故障诊断领域，开发设计出一个集数据采集，信号处理，故障诊断和网路通讯等功能于一体的在线故障诊断仪，从而提高了现场诊断的能力，简化了传统诊断系统的结构，改善了系统的稳定性。