Machine learning as an online diagnostic tool for proton exchange membrane fuel cells

Proton exchange membrane fuel cells are considered a promising power supply system with high efficiency and zero emissions. They typically work within a relatively narrow range of temperature and humidity to achieve optimal performance; however, this makes the system difficult to control, leading to faults and accelerated degradation. Two main approaches can be used for diagnosis, limited data input which provides an unintrusive, rapid but limited analysis, or advanced characterisation that provides a more accurate diagnosis but often requires invasive or slow measurements. To provide an accurate diagnosis with rapid data acquisition, machine learning methods have shown great potential. However, there is a broad approach to the diagnostic algorithms and signals used in the field. This article provides a critical view of the current approaches and suggests recommendations for future methodologies of machine learning in fuel cell diagnostic applications.     

Heart Failure and PAHs,OHPAHs, and Trace Elements Levels in Human Serum: Results from a Preliminary Pilot Study in Greek Population and the Possible Impact of Air Pollution

Cardiovascular diseases (CVDs) have been associated with environmental pollutants. The scope of this study is to assess any potential relation of polycyclic aromatic hydrocarbons (PAHs), their hydroxylated derivatives, and trace elements with heart failure via their direct determination in human serum of Greek citizens residing in different areas. Therefore, we analyzed 131 samples including cases (heart failure patients) and controls (healthy donors), and the respective demographic data were collected. Significantly higher concentrations (p < 0.05) were observed in cases’ serum regarding most of the examined PAHs and their derivatives with phenanthrene, fluorene, and fluoranthene being the most abundant (median of >50 μg L⁻¹). Among the examined trace elements, As, Cd, Cu, Hg, Ni, and Pb were measured at statistically higher concentrations (p < 0.05) in cases’ samples, with only Cr being significantly higher in controls. The potential impact of environmental factors such as smoking and area of residence has been evaluated. Specific PAHs and trace elements could be possibly related with heart failure development. Atmospheric degradation and smoking habit appeared to have a significant impact on the analytes’ serum concentrations. PCA–logistic regression analysis could possibly reveal common mechanisms among the analytes enhancing the hypothesis that they may pose a significant risk for CVD development.     

高压直流断路器测控装置的设计与试验分析

直流输(配)电是目前电网发展的趋势,而高压直流断路器的研发是影响其发展的关键技术之一。测量与控制装置的研究是高压直流断路器的研发要点。采用高压电力电子器件,DSP+FPGA相结合的控制方式,对其测控装置进行研究分析,提出了区间阈值的控制方法及相应的控制时序,从高压直流输电、能源多样化的发展需求入手,将机械开关和固态开光相结合,设计了一套混合式高压直流断路器样机。通过对样机进行合分闸试验及故障分闸试验,证明了所研制的高压直流断路器样机具有分闸动作快,限弧能力强,动作一致性好等优点。     

基于Flowmaster的飞机燃油供油系统仿真研究

飞机燃油系统作为飞机不可或缺的功能和保障系统,对飞机的安全性有着重大的影响。针对飞机燃油供油系统的故障诊断问题,利用流体仿真软件Flowmaster建立了供油系统模型。仿真了飞机燃油系统增压泵供油、交输供油、重力供油的工作情况。分析了飞机燃油供油系统的故障模式,仿真了在几种典型故障模型下的工作情况,并对仿真结果进行了分析。结果表明利用Flowmaster所建立的供油系统模型能有效地仿真飞机燃油系统正常与故障工作情况,为飞机燃油系统的故障诊断打下了基础。     

SOFA评分联合超声对脓毒症合并AKI患者的预后价值分析

本文主要探究SOFA评分联合超声对脓毒症合并急性肾损伤(AKI)患者的预后价值。选取2017年1月~2019年1月本院脓毒症合并AKI患者50例作为观察组,并分为AKI 1、2、3期,将同期入院的50例脓毒症患者作为对照组,两组患者均采用SOFA评分联合超声进行预后评估。结果发现,观察组PDU评分低于对照组(P<0.05),RI值、SOFA评分高于对照组(P<0.05);不同分期的3组间的PDU评分、SOFA评分不同,随着AKI分期的增加,PDU评分降低、SOFA评分增加(P<0.05),但3组间的RI值并无不同(P>0.05);50例脓毒症合并AKI患者发生院内死亡率为44.00%。经单因素分析发现,年龄、机械通气时间、ICU住院时间、AKI分期、脓毒症休克、SOFA评分、PDU评分为影响患者预后不佳的因素(P<0.05);AKI3期、发生脓毒性休克、SOFA评分、PDU评分是脓毒症合并AKI患者预后不佳的独立因素(P<0.05),ROC曲线下面积(AUC)越大,对预后的预测效能越好,当AUC>70.00%时具有临床价值。两者联合显著高于单独应用SOFA评分(AUC=74.28%)或PDU评分(P<0.001)。上述结果说明,脓毒症合并AKI患者采用SOFA评分联合超声用于评估患者的预后,优于单独采用SOFA评分或超声,两者联合的预测价值更大。     

修理设备可更换且修理有延迟的两不同型部件并联可修系统

假定部件的寿命服从指数分布,其修理延迟时间和修理时间均服从一般分布,并且修理设备的寿命服从指数分布,其更换时间服从一般分布,利用马尔可夫更新过程理论和一种新的分解方法,研究了修理设备可更换且修理有延迟的两不同型部件并联可修系统,求得了系统和修理设备有关可靠性指标的一系列结果.     

复合材料失效判据有效性验证方法研究

为系统地验证复合材料失效判据计算精度和有效性范围,给出了4种材料体系、6种铺层形式的层合板在单轴、双轴载荷下的失效试验数据,用以评估复合材料失效判据在单向层合板失效包线、多向层合板初始失效包线、多向层合板最终失效包线、层合板变形及层合板的破坏特性等五个方面的预测能力。并根据验证方法和有效性评估策略对失效判据计算精度进行量化考核,给出了失效判据在五个层面上的计算精度。     

基于持续同调的边坡土颗粒应力链失稳分析

采用颗粒离散元方法和持续同调理论,研究了内排土场堆叠至不同高度时的边坡稳定性。为便于研究,现采用一水平金属板向下施加压力,代替不同厚度土层的重力荷载,对边坡在竖向荷载作用下的失稳破坏过程进行了颗粒离散元模拟。研究了二维边坡土颗粒速度总矢量、边坡失稳破坏时滑移开裂面的角度以及边坡坡顶y方向的平均速度等宏观响应过程,并构建了自然堆积下边坡堆积体颗粒的法向力链无向网络模型。最后,用持续同调方法对边坡坡顶颗粒接触力链网络的拓扑特征进行分析,获得条码图,建立了岩体结构持续同调特征与失稳演化的关系。本文为研究边坡失稳拓扑识别提供了一种新方法,从而可以有效预测边坡的失稳破坏。     

多层网络级联失效的预防和恢复策略概述

现实生活中,与国计民生密切相关的基础设施网络大多不是独立存在的,而是彼此之间相互联系或依赖的,于是用于研究这些系统的多层网络模型随之产生.多层网络中的节点在失效或者遭受攻击后会因"层内"和"层间"的相互作用而产生级联效应,从而使得失效能够在网络层内和层间反复传播并使得失效规模逐步放大.因此,多层网络比单个网络更加脆弱.多层网络级联失效产生的影响和损失往往是非常巨大的,所以对多层网络级联失效的预防和恢复的研究具有重大意义.就多层网络级联失效的预防而言,主要包含故障检测,保护重要节点,改变网络耦合机制和节点备份等策略.就多层网络发生级联失效后的恢复策略而言,主要包含共同边界节点恢复、空闲连边恢复、加边恢复、重要节点优先恢复、更改拓扑结构、局域攻击修复、自适应边修复等策略.