信号交叉口延误计算模型研究

利用深圳市主要信号交叉口延误实际调查结果,结合美国2000年版HCM延误模型计算结果,应用最小二乘法对模型进行修正,建立与深圳市交叉口延误相符合的延误计算模型,为深圳市信号交叉口和道路系统的分析评价提供了理论依据.     

分布式入侵检测模型研究

提出了分布级别的概念对分布式入侵检测系统进行分类，并引入信息抽象级别对入侵检测中审计数据所经历的逻辑抽象层次进行表述．在对现有的层次检测模型和协作检测模型的优点和缺陷进行详细分析之后，提出了一种用于分布式入侵检测系统的层次化协作模型(HCM)，并完成了相应的原型系统．该模型可以有效地综合两种现有模型的优点，在保证结点可控性和检测效率的同时提高系统的容错性和协作能力．     

自适应径向基神经网络及其应用

提出一种基于硬C均值算法的自适应RBF神经网络。该算法根据网络训练误差的变化,在隐层到输出层的权值修改过程中,对学习步长进行自适应调节;对通常采用的基函数宽度的计算方法作了改进;对于硬C均值算法出现的死节点,则在程序运行中自动进行删除。利用该改进的自适应RBF网络进行某合成氨装置的氢氮比预测,网络计算误差小、收敛迅速、结果令人满意,表明网络具有良好的性能。     

基于直方图加权的半模糊聚类

通过分析影像数据的特点,利用直方图的统计特性,结合HCM收敛速度快的优点,提出了一种基于直方图加权的半模糊化的聚类算法,此方法结合了全局与局部信息,提高了聚类的速度,改善了聚类的效果;采用Lena和脑影像实验与传统算法作比较证明了该算法的效果更好,并对一副97 658k的影像进行处理,证明了该算法效率高。     

基于模糊聚类的半自动跟踪技术研究

现有半自动跟踪系统将操作员等效为准线性环节,增大了训练的难度和成本,不适应复杂战场环境下多个快速机动目标的选取和跟踪.利用特征提取方法得到视场的“潜在目标”集,引入模糊聚类方法,通过对操作响应进行分类,建立操控意图与目标运动特征之间的关系模型.通过充分的仿真试验,对比研究了跟踪平稳性、快速性、准确性等指标,结果说明该方法能够有效提升整体跟踪性能.     

信号交叉口HCM2000法延误计算修正模型研究

判定信号交叉口信号配时方案设计是否合理的重要依据是进行延误分析与定量计算。美国HCM2000延误算法是国际上很经典的一种延误计算模型,但因不同地域交通条件的变化,在不同地域应用时延误计算结果与实测值会有不同程度的误差。采用点样本法在对哈尔滨市部分信号交叉口进行延误调查的基础上,应用HCM2000算法对所调查各交叉口进行延误计算,通过将延误理论计算值与实测值比较分析发现HCM2000算法与实测值有较大差异。而后文中基于建立与哈尔滨市交通情况相适应的延误计算模型的思想,使用SPSS软件建立HCM2000算法延误计算修正模型,为哈尔滨市信号交叉口控制研究和道路系统的分析评价中延误的分析与计算提供一种适用的延误计算模型。     

基于HCM聚类的连续域模糊关联算法

针对粗糙集对于连续域属性决策表的处理能力差以及不容易获得模糊集之间关系等问题,提出一种基于连续型属性模糊关联规则约简算法。该算法引入三角隶属度函数将连续属性值转化为模糊值,并使用硬C均值聚类方法获得数据集之间关系,采用遗传算法优化该模型。仿真结果验证了该模型的有效性。     

Influence of long-term service on microstructure,mechanical properties,and service life of HCM12A steel

The investigated material was a pipe measuring φ38?×?6.3, made of HCM12A steel used for coils of boiler superheaters at supercritical and ultra-supercritical parameters of steam. The article presents the results of research on mechanical properties and microstructure of the examined pipe material in the as-received condition and after 100?000?h of service at the temperature of 585°C. Creep tests were carried out for the material serviced for 100?000?h to determine the available residual life. The influence of changes on mechanical properties and microstructure degradation on the loss of creep resistance was assessed compared with the as-received condition. Performed research has shown that the examined steel after 100?000?h of service fulfills the requirements set for this steel in the as-received condition within the range of strength properties at room and elevated temperature. The insignificant loss of life is confirmed by the results of creep test.     

一种新的聚类分析算法

给出了一种新的无监督聚类算法,但这种算法并非是基于目标函数的聚类算法,而是对数据直接设计一种迭代运算,以使数据在保持类特征的情况下进行重新组合最终达到分类的目的。通过对一类数据的实验表明,该算法在无监督给出类数方面具有较好的鲁棒性;另外,该算法在数据的准确归类、无监督聚类、确定性,以及对特殊类分布的适用性等方面均优于ＨＣＭ和ＦＣＭ算法,     

The Time Has Come to Explore Plasma Biomarkers in Genetic Cardiomyopathies

For patients with hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM) or arrhythmogenic cardiomyopathy (ACM), screening for pathogenic variants has become standard clinical practice. Genetic cascade screening also allows the identification of relatives that carry the same mutation as the proband, but disease onset and severity in mutation carriers often remains uncertain. Early detection of disease onset may allow timely treatment before irreversible changes are present. Although plasma biomarkers may aid in the prediction of disease onset, monitoring relies predominantly on identifying early clinical symptoms, on imaging techniques like echocardiography (Echo) and cardiac magnetic resonance imaging (CMR), and on (ambulatory) electrocardiography (electrocardiograms (ECGs)). In contrast to most other cardiac diseases, which are explained by a combination of risk factors and comorbidities, genetic cardiomyopathies have a clear primary genetically defined cardiac background. Cardiomyopathy cohorts could therefore have excellent value in biomarker studies and in distinguishing biomarkers related to the primary cardiac disease from those related to extracardiac, secondary organ dysfunction. Despite this advantage, biomarker investigations in cardiomyopathies are still limited, most likely due to the limited number of carriers in the past. Here, we discuss not only the potential use of established plasma biomarkers, including natriuretic peptides and troponins, but also the use of novel biomarkers, such as cardiac autoantibodies in genetic cardiomyopathy, and discuss how we can gauge biomarker studies in cardiomyopathy cohorts for heart failure at large.