Microfluidic systems in clinical diagnosis

The use of microfluidic devices is highly attractive in the field of biomedical and clinical assessments, as their portability and fast response time have become crucial in providing opportune therapeutic treatments to patients. The applications of microfluidics in clinical diagnosis and point-of-care devices are continuously growing. The present review article discusses three main fields where miniaturized devices are successfully employed in clinical applications. The quantification of ions, sugars, and small metabolites is examined considering the analysis of bodily fluids samples and the quantification of this type of analytes employing real-time wearable devices. The discussion covers the level of maturity that the devices have reached as well as cost-effectiveness. The analysis of proteins with clinical relevance is presented and organized by the function of the proteins. The last section covers devices that can perform single-cell metabolomic and proteomic assessments. Each section discusses several strategically selected recent reports on microfluidic devices successfully employed for clinical assessments, to provide the reader with a wide overview of the plethora of novel systems and microdevices developed in the last 5 years. In each section, the novel aspects and main contributions of each reviewed report are highlighted. Finally, the conclusions and future outlook section present a summary and speculate on the future direction of the field of miniaturized devices for clinical applications.     

Detection of telomerase activity by combination of telomeric repeat amplification protocol and electrochemiluminescence assay

A highly sensitive telomerase detection method that combines telomeric repeat amplification protocol （TRAP） and magnetic beads based electrochemiluminescence （ECL） assay has been developed. Briefly, telomerase recognizes biotinylated telomerase synthesis primer （B-TS） and synthesizes extension products, which then serve as the templates for PCR amplification using B-TS as the forward primer and tris-（2′2′-bipyridyl） ruthenium （TBR） labeled ACX （TBR-ACX） as the reversed primer. The amplified product is captured on streptavidin-coated paramagnetic beads and detected by ECL. Telomerase positive HeLa cells were used to validate the feasibility of the method. The experimental results showed down to 10 cancer cells can be detected easily. The method is a useful tool for telomerase activity analysis due to its sensitivity, rapidity, safety, high throughput, and low cost. It can be used for screening a large amount of clinical samples.     

Stabilization of active fault‐tolerant control systems by uncertain nonhomogeneous markovian jump models

This article investigates the stabilization and control problems for a general active fault‐tolerant control system (AFTCS) in a stochastic framework. The novelty of the research lies in utilizing uncertain nonhomogeneous Markovian structures to take account for the imperfect fault detection and diagnosis (FDD) algorithms of the AFTCS. The underlying AFTCS is supposed to be modeled by two random processes of Markov type; one characterizing the system fault process and the other describing the FDD process. It is assumed that the FDD algorithm is imperfect and provides inaccurate Markovian parameters for the FDD process. Specifically, it provides uncertain transition rates (TRs); the TRs that lie in an interval without any particular structures. This framework is more consistent with real‐world applications to accommodate different types of faults. It is more general than the previously developed AFTCSs because of eliminating the need for an accurate estimation of the fault process. To solve the stabilizability and the controller design problems of this AFTCS, the whole system is viewed as an uncertain nonhomogeneous Markovian jump linear system (NHMJLS) with time‐varying and uncertain specifications. Based on the multiple and stochastic Lyapunov function for the NHMJLS, first a sufficient condition is obtained to analyze the system stabilizability and then, the controller gains are synthesized. Unlike the previous fault‐tolerant controllers, the proposed robust controller only needs to access the FDD process, besides it is easily obtainable through the existing optimization techniques. It is successfully tested on a practical inverted pendulum controlled by a fault‐prone DC motor. © 2016 Wiley Periodicals, Inc. Complexity 21: 318–329, 2016     

中国上市公司总资产与营业收人关系的实证分析

企业将资产运用于生产经营活动,并由此赚取更多的资产,即产生公司的收入.因此企业资产与收入之间必定存在一定的相关关系.在对上市公司总资产与营业收入进行一般线性拟合的基础之上,采用分位回归模型对上市公司的总资产与营业收入的关系进行深入剖析.结果表明,传统的线性模型只能揭示出总资产与营业收入呈正相关关系,而分位回归方法能更好地看出,高分位点营业收入的企业在提高一定总资产时,会更能促进营业收入的增长.由于收集到的数据中存在离群点,在第5节讨论了线性分位回归模型的统计诊断,类比于一般线性模型的R square得到不同分位点上的R square.通过删除离群点的处理,得出分位回归模型比一般线性模型更加稳健,数据在高分位点的拟合效果更好一些.     

故障树模块化在装甲车辆灭火系统故障诊断中的应用

为了实现装甲装备灭火系统故障的快速诊断,提出了一种故障树模块化分析方法;对灭火系统故障树进行深度优先最左遍历,并记录遍历过程,按照遍历顺序对故障树中的每个事件进行标定,并将灭火系统故障树划分为相互独立的模块,依据划分的模块可以通过故障现象对模块内的故障进行排查及修复;实验结果分析表明,该方法可以快速修复模块故障,恢复系统功能,简化了以往对灭火系统所有子事件遍历查错的繁琐过程。该方法同样可以计算故障模块的失效概率,并可以实现故障模块的整体更换,恢复系统性能;证明了故障树模块化方法在灭火系统故障诊断中具有较高的效率,简化了灭火系统诊断流程,在装甲车辆其他系统故障诊断中具有借鉴作用,符合现代作战对于装备保障的需求。     

基于极限学习机与规则推理的NPC三电平逆变器二级故障诊断方法

针对NPC三电平逆变器故障诊断问题,提出一种基于极限学习机与规则推理的二级故障诊断方法。分析了依据输出电流诊断故障的可区分性,以及故障模式的分类。然后对输出电流提取故障特征,并采用极限学习机完成故障初级分类。对于初级分类结果为电流不可区分故障情况,再根据桥臂电压信息运用规则推理法实现故障二级精确诊断。诊断实验表明,该方法能够实现NPC三电平的多模式故障诊断,且故障诊断方法简单、定位精确、快速、鲁棒性强。     

基于贝叶斯网络推理的起落架系统故障诊断技术研究

民机起落架系统结构复杂,是典型的故障多发系统,实际诊断过程主要依赖于排故手册流程和工程经验积累,存在诸多不确定性因素。贝叶斯网络是用有向无环图的形式表达变量间因果关联关系,可以充分利用专家知识和试验信息进行基于概率的统计推断,适于处理复杂系统的不确定性问题。通过深入分析某型民机起落架技术资料,建立了基于贝叶斯网络的起落架系统诊断架构,结合专家知识和维护经验提出了基于贝叶斯网络的起落架系统故障诊断方法,并给出了网络推理流程,提升了起落架系统故障诊断效率和精度。     

无人机健康诊断知识库建立研究

针对无人机健康诊断系统的设计,研究了健康诊断专家知识库的建立方法。介绍了基于规则的专家知识库的含义及其对无人机设计的要求,阐述了故障树与故障模式判据表的形成方法。故障树分析基于故障因果逻辑,逐层找出故障事件的原因,保证专家知识逻辑上的完备性。故障模式判据表将抽象的专家知识具体化为多个能够在工程上应用的要素。提出了一种故障树分析与故障判据规则相结合的建立健康诊断专家知识库的方法。从某型大气数据系统的组成及原理出发,以大气数据系统真空速失效为顶事件,构造了故障树和故障模式判据表。结果表明,结合故障树和故障模式判据表格构造的专家知识库清晰、简洁,具有很高的工程实用价值,能够应用于无人机健康诊断系统的设计。     

基于主元分析与动态时间弯曲的故障诊断方法及应用研究

轴承是工程实际中常用而又极易损坏的部件,特别是对其早期微弱响应的辨识,具有重要的社会价值和意义。为提高运转轴承的安全可靠性和可维护性,提出了基于主元分析与动态时间弯曲距离的故障诊断方法,它可以准确对早期微弱动态响应辨识、诊断。该方法首先将典型故障样本信号与待测信号小波去噪并EMD分解,并对若干固有模态分量主元分析求取主元,然后对主元分量进行分析,获得相关特征值组成特征向量,计算待测信号与已知故障样本信号特征向量的弯曲距离,弯曲距离越小表明两信号越相似,从而辨识故障。此外,还可将其应用于转子、碰磨、齿轮故障诊断中,工程应用实例表明该方法可以准确故障分类,高效故障诊断。     

基于气象因素的集中供热系统热负荷预测研究

为了解决直升机旋翼伺服作动器维修保障过程中训练资源有限,拆装难度高,故障率较高且难以确定故障点等难题,在直升机虚拟维修训练系统(VMTS, Virtual Maintenance Training System)的基础上,提出了应用虚拟仪器检测,确定故障点的直升机故障诊断可视化设计方法。分析了故障诊断过程中仪器、工具、环境等对故障诊断的影响,实现了动态的建立仿真模型和数据处理。通过虚拟仪器和推理机的功能配合,解决了直升机维护中复杂多变的故障诊断难题,形成一个逻辑推理能力强的故障诊断系统。最后以直升机旋翼伺服作动器故障诊断为例,验证该方法的可行性。