基于支持向量机的多目标分类和识别

支持向量机(SVM)算法广泛应用于模式识别等领域,但是SVM最初是针对二类别分类提出,在多分类识别中稍显逊色。对将SVM由二分类扩展到多分类的算法进行了研究,发现有向无环图(DAG-SVM)是其中用的最多的算法之一。因此,针对军事领域图像的多目标分类,选择有向无环图算法来实现军事图像中单兵、装甲、低空等多目标的分类识别。     

蓄电池注水装置水位控制原理研究与实现

研究设计了由电磁式液位传感器、隔离开关、单片机控制器及报警系统组成的潜艇蓄电池注水装置,确保了各蓄电池液面误差不大于5mm.给出了该装置结构及工作原理,分析了液位传感器和单片机控制电路.经使用证明,该装置对控制注水高度及缩短注水时间有良好效果.     

基于可调光纤F-P滤波器解调的分布式光纤光栅传感系统

提出并实现了一种可满足工程实用要求的波分复用光纤光栅传感网络解决方案,系统通过可调谐光纤F-P滤波器的连续扫描实现波长信号的解调.该传感系统扫描带宽50 nm,单点工作带宽5 nm,对一般应用系统每根单纤可设20-30个点,系统最高扫描频率200 Hz,分辨率5 pm(约5με或0.5℃).     

电发火系统感应电流测试方法研究

对电发火系统在电磁环境下的感应电流测量采用一种新型的白光干涉型光纤测温仪。通过测量电发火系统中电火工品桥丝在电磁环境下温度的变化,利用桥丝温度和感应电流的关系,测量出桥丝上的感应电流,进而对弹箭系统进行电磁环境效应评估。     

发动机传感器数据证实的贝叶斯信度网络方法

针对发动机及其部件试验传感器数据证实的多源证据融合问题,描述了建立贝叶斯信度网络的方法,给出了传感器状态和检验关系式不确定性信息表达方法,发展了自动建立贝叶斯信度网络、计算可信度概率及更新网络的算法;给出了贝叶斯信度网络方法在模型发动机上的应用示例。     

高速磁浮列车气隙磁场测量系统的研究

介绍了高速磁浮列车气隙磁场测量系统研制中的几项关键技术,提出了三维磁敏传感器的设计方法和基于霍尔元件的电压比磁场测量法,并设计了传感器的高精度运动控制系统。所研制的系统具有较高的测磁精度和定位精度,实现了高速磁浮列车气隙磁场的自动测量。     

Optimal source placement for point coverage in active multistatic sonar networks

The allocation of underwater sensors for tracking, localization, and surveillance purposes is a fundamental problem in anti-submarine warfare. Inexpensive passive receivers have been heavily utilized in recent years; however, modern submarines are increasingly quiet and difficult to detect with receivers alone. Recently, the idea of deploying noncollocated sources and receivers has emerged as a promising alternative to purely passive sensor fields and to traditional sonar fields composed of collocated sources and receivers. Such a multistatic sonar network carries a number of advantages, but it also brings increased system complexity resulting from its unusual coverage patterns. In this work, we study the problem of optimally positioning active multistatic sonar sources for a point coverage application where all receivers and points of interest are fixed and stationary. Using a definite range sensor model, we formulate exact methods and approximation algorithms for this problem and compare these algorithms via computational experiments. We also examine the performance of these algorithms on a discrete approximation of a continuous area coverage problem and find that they offer a significant improvement over two types of random sensor deployment.     

涡轮泵状态监控及传感器故障识别的新异类检测方法

为了在故障样本稀缺、故障模式不完备的情况下监控涡轮泵状态,并剔除传感器失效故障造成的虚警,提出用于涡轮泵状态监控及传感器故障识别的单类支持向量机新异类检测方法。该方法以正常状态为目标类构建单类支持向量机检测器,用于检测涡轮泵是否出现异常;以传感器故障为目标类构建单类支持向量机检测器,用于判断检测到的异常是否属于传感器故障。对涡轮泵试车数据的分析结果表明了该方法的有效性。     

装备预测与健康管理体系结构及关键技术

结合预测与健康管理技术（PHM）在降低装备保障维修费用方面的明显优势,介绍了预测与健康管理技术的内涵及特点,分析了PHM系统的开放式流程,并将该开放式流程应用于装备的故障诊断和预测中,提出一种基于PHM的预测与健康管理体系框架。在此基础上,对诊断与预测技术以及传感器配置算法等关键技术进行了详细阐述,介绍了采用的诊断和预测模型及算法。     

Online process mean estimation using L1 norm exponential smoothing

A basic assumption in process mean estimation is that all process data are clean. However, many sensor system measurements are often corrupted with outliers. Outliers are observations that do not follow the statistical distribution of the bulk of the data and consequently may lead to erroneous results with respect to statistical analysis and process control. Robust estimators of the current process mean are crucial to outlier detection, data cleaning, process monitoring, and other process features. This article proposes an outlier‐resistant mean estimator based on the L₁ norm exponential smoothing (L₁‐ES) method. The L₁‐ES statistic is essentially model‐free and demonstrably superior to existing estimators. It has the following advantages: (1) it captures process dynamics (e.g., autocorrelation), (2) it is resistant to outliers, and (3) it is easy to implement. © 2009 Wiley Periodicals, Inc. Naval Research Logistics 2009