共查询到20条相似文献,搜索用时 62 毫秒
1.
2.
直接将入侵检测算法应用在粗糙数据上,其入侵检测分析的效率非常低.为解决该问题,提出了一种基于主成分分析的入侵检测方法.该方法通过提取网络连接中的相关信息,对它进行解码,并将解码的网络连接记录与已知的网络连接记录数据进行比较,发现记录中的变化和连接记录分布的主成分,最后将机器学习方法和主成分分析方法结合实现入侵检测.实验结果表明该方法应用到各种不同KDD99入侵检测数据集中可以有效减少学习时间、降低各种数据集的表示空间,提高入侵检测效率. 相似文献
3.
4.
5.
提出了一种新的虹膜特征提取与识别方法,该方法利用核主成分分析(KPCA)在高维空间具有较强的特征选择能力来提取虹膜图像的纹理特征。采用了一种距离度量和支持向量机相结合的两级分类方法,前级采用欧式距离来度量图像间的相似性,若符合条件,给出分类结果,否则拒绝,并转入后一级分类器——支持向量机分类,以减少进入支持向量机的样本数目,该组合分类方法充分利用了支持向量机识别率高和距离度量速度快的优点。实验结果表明,该方法提高了虹膜识别率,是一种有效的虹膜识别方法。 相似文献
6.
运用主成分分析方法,经过矩阵变换,降低维数以提取故障信息的主要特征,实现对发电设备故障数据二维可视化,之后使用支持向量机来获取最优分类面进行诊断分析。仿真数据表明,在有用信息量损失较小的前提下,该方法可以提炼故障信息而不影响分析结果,是解决多变量高维复杂系统简便有效的一种识别方法。 相似文献
7.
8.
9.
基于结构动态响应时程和主成分分析技术对结构损伤识别和结构健康状态在线监测进行研究。采用相关性分析方法,提出了一种新的损伤敏感指标——主成分置信度。首先分别对各测点实测时域响应数据进行分段、重组并形成矩阵形式,再进行标准化处理;其次,利用主成分分析提取各测点的主成分,对主成分进行相关性分析,并计算主成分置信度;最后,通过若干结构损伤仿真计算和试验验证了该方法的正确性和有效性。该方法仅需提取结构基准状态和待测状态的时域响应特征,即可进行损伤判别,计算量小且不限于结构型式,因此,非常有利于结构在线监测。 相似文献
10.
11.
12.
13.
14.
Saruar Alam Goo‐Rak Kwon The Alzheimer's Disease Neuroimaging Initiative 《International journal of imaging systems and technology》2017,27(2):133-143
Early diagnosis of Alzheimer disease (AD) and mild cognitive impairment (MCI) is always useful. Preventive measures might have an impact on reducing AD risk factors. Structural magnetic resonance (MR) imaging, one of the vital sensitive biomarkers for cerebral atrophy in the brain, is used to extract volumetric feature by FreeSurfer and the CIVET toolbox. All of the structural magnetic resonance imaging (s‐MRI) data that we used were downloaded from the Alzheimer's disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu) of imaging data. This novel approach is applied for the diagnosis of AD and MCI from healthy controls (HCs) combining extracted features with the MMSE (mini‐mental state examination) scores, applying a two sample t‐test to select a subset of features. The subset of features is fed to kernel principal component analysis (KPCA) module to project data onto the reduced principal component coefficients at higher dimensional kernel space to increase the linear separability. Then, the kernel PCA coefficients are projected into the more efficient linear discriminant space using linear discriminant analysis. A multi‐kernel learning support vector machine (SVM) is used on newly projected data for stratification of AD and MCI from HCs. Using this approach, we obtain 93.85% classification accuracy when detecting AD from HCs for segmented volumetric features (using FreeSurfer) with high sensitivity and specificity. When distinguishing MCI from HCs and AD using volumetric features after subcortical segmentation, the detection rate reaches 86.54% and 75.12%, respectively. 相似文献
15.
糜克勤 《材料科学技术学报》1990,6(5):333-337
Based on thermodynamic calculation,oper-ating data and experiments,Si transmission in B.F.is described in this paper.The Si in hot metal is con-trolled by both reduction and reoxidation.Equilib-rium calculation gives a tapping value and the pro-file of Si content of the hot metal in B.F.The firstimportant factor to control Si content is reactiontemperature.The coke reactivity and solution lossare more sensitive to in-situ coke reduction and Sicontent. 相似文献
16.
基于H∞鲁棒控制原理,提出了结构主动控制系统故障检测器的设计方法,形成了结构主动控制系统的故障诊断与隔离技术(fault detection and isolation,FDI)。首先根据H∞控制目标提出了FDI系统的模型,将故障信号的检测问题转化为一个鲁棒控制问题,应用线性矩阵不等式(linear matrix inequality,LMI)算法解出H∞最优控制器,即故障检测器。通过一个三层框架结构的数值算例验证了设计的有效性。该研究工作为结构主动控制系统的健康检测提供了参考,为进一步进行容错控制的研究奠定了基础。 相似文献
17.
Changkyu Lee Sang Wook Choi In-Beum Lee 《Chemometrics and Intelligent Laboratory Systems》2004,70(2):165-178
Principal component analysis (PCA) is widely employed as a multivariate statistical method for fault detection, isolation and diagnosis in chemical processes. Previously, PCA has been successfully used to identify faulty sensors under normal static operating conditions. In this paper, we extend the reconstruction-based sensor fault isolation method proposed by Dunia et al. to dynamic processes. We develop a new method for identifying and isolating sensor faults in an inherent dynamic system. First, we describe how to reconstruct noisy or faulty measurements in dynamic processes. The reconstructed measurements are obtained by simple iterative optimization based on the correlation structure of the time-lagged data set. Then, based on the sensor validity index (SVI) approach developed by Dunia et al., we propose an SVI for fault isolation in dynamic processes. The proposed method was applied to sensor fault isolation in two strongly dynamic systems: a simulated 4×4 dynamic process and a simulated wastewater treatment process (WWTP). In these experiments, the proposed sensor fault identification method correctly and rapidly identified the faulty sensor; in contrast, the traditional PCA-based sensor fault isolation approach showed unsatisfactory results when applied to the same systems. 相似文献
18.
The activity of MnO in blast furnace-typeslags CaO-SiO_2,CaO-SiO_2-Al_2O_3 andCaO-SiO_2-Al_2O_3-MgO at 1500℃ has been measuredin a graphite crucible with Ag and Sn as solventmetals.The agreement between the values ofγMnO obtained from the experimental resultsusing Ag and Sn separately as the solvent metalis excellent.The major errors in the previousstudies have proved to be the failure to takeinto account the exchange reactions involvingSi and Mn and insufficient time for theestablishment of equilibrium between slag andsolvent metal.A new method is suggested for determiningthe activities of elements in metallic solutions.The method involves the separate equilibrationof two metallic alloys,one of which containsthe component of interest at a known activity,with a single slag. 相似文献
19.
20.
炼铁高炉炉衬爆破拆除 总被引:2,自引:0,他引:2
本文介绍靶钢炼铁厂11^#高炉大修工程中爆破拆除高炉炉衬的过程。采用爆破方法先破坏炉衬环型砌体结构,再爆破冷却壁之上衬砖,使衬砖局部悬空,在重力作用下脱落,达到安全,快速拆除衬砖的目的。 相似文献