机动车车载尾气检测技术研究综述

车载尾气检测技术的出现和发展为机动车尾气的相关研究提供了一种较为准确、便捷、实时反映城市路网排放的测试方法。本文详细介绍了车载尾气检测技术的发展历程和原理,综述了该技术在行驶周期开发、尾气排放特性和规律研究、建立机动车尾气排放数据库和建立城市路网排放清单等领域的应用现状,并通过比较该技术在国内、外的应用差距,提出了在绿色交通领域充分利用该技术进行机动车污染评价与控制的合理建议。     

压力脉冲波法天然气管道堵塞检测研究*

为快速检测出天然气管道堵塞位置、长度及堵塞程度,解决天然气输运管道中堵塞定位难题,探讨压力脉冲波法管道堵塞检测的可行性与效果。搭建长220 m的压力脉冲波堵塞检测实验系统,进行多组连续性长堵塞检测实验和水堵检测实验。结果表明:连续性长堵塞检测实验中堵塞前沿反射信号为负压波,堵塞后沿信号为正压波,且波形清晰可辩,其多组实验的堵塞位置、堵塞长度及堵塞率的平均检测误差分别为0.23%,0.75%和2.67%;水堵反射信号清晰可辨但波形复杂,其多组实验的堵塞位置、堵塞长度及堵塞率的平均检测误差分别为0.78%,9.61%和43.28%。实验结果有效地证实了压力脉冲法的可靠性与实用性,并为压力脉冲波法堵塞检测现场应用提供指导。     

基于自适应局部斥力与归一化面积损失的工程车辆目标检测*

为加强施工场景下的工程车辆安全监管,针对施工场景下工程车辆易互相遮挡、局部特征与全局特征相似以及场景环境复杂的问题,提出一种基于自适应局部斥力与归一化面积损失的工程车辆目标检测算法。其中自适应局部斥力损失可使待检测工程车辆与其他工程车辆目标框相排斥;归一化面积损失使网络学习集中在面积具有相对较大预测误差的工程车辆上;并结合聚类算法设定更适合工程车辆的锚框。研究结果表明:算法可实现在困难场景下对压路机、挖掘机、装载机3类工程车辆的快速准确检测与识别,具有较高的工程应用价值。     

基于聚类算法的ERT污染区域识别方法

本文提出将聚类算法引入到ERT监测系统中，采用K均值（K-means）聚类、模糊C均值算法（FCM）以及混合高斯模型（GMM）3种常用聚类算法对ERT检测结果进行污染区域识别，通过一个数值模型分析了3种算法的识别效果.研究结果表明当污染区域与背景土壤的电阻率区分度较大时（电阻率差异性大于30%），采用3种聚类算法都可以识别出污染区域，K-means和FCM的识别效果优于GMM算法.最后，给出一个实际场地调查的应用案例.     

Rapid Increases and Time‐Lagged Declines in Amphibian Occupancy after Wildfire

Climate change is expected to increase the frequency and severity of drought and wildfire. Aquatic and moisture‐sensitive species, such as amphibians, may be particularly vulnerable to these modified disturbance regimes because large wildfires often occur during extended droughts and thus may compound environmental threats. However, understanding of the effects of wildfires on amphibians in forests with long fire‐return intervals is limited. Numerous stand‐replacing wildfires have occurred since 1988 in Glacier National Park (Montana, U.S.A.), where we have conducted long‐term monitoring of amphibians. We measured responses of 3 amphibian species to fires of different sizes, severity, and age in a small geographic area with uniform management. We used data from wetlands associated with 6 wildfires that burned between 1988 and 2003 to evaluate whether burn extent and severity and interactions between wildfire and wetland isolation affected the distribution of breeding populations. We measured responses with models that accounted for imperfect detection to estimate occupancy during prefire (0–4 years) and different postfire recovery periods. For the long‐toed salamander (Ambystoma macrodactylum) and Columbia spotted frog (Rana luteiventris), occupancy was not affected for 6 years after wildfire. But 7–21 years after wildfire, occupancy for both species decreased ≥25% in areas where >50% of the forest within 500 m of wetlands burned. In contrast, occupancy of the boreal toad (Anaxyrus boreas) tripled in the 3 years after low‐elevation forests burned. This increase in occupancy was followed by a gradual decline. Our results show that accounting for magnitude of change and time lags is critical to understanding population dynamics of amphibians after large disturbances. Our results also inform understanding of the potential threat of increases in wildfire frequency or severity to amphibians in the region. Incrementos Rápidos y Declinaciones Desfasadas en la Ocupación de Anfibios Después de un Incendio     

Anti-predatory vigilance and the limits to collective detection: visual and spatial separation between foragers

Collective detection concerns the idea that all members of a socially feeding group are alerted to an attack as long as at least one group member detects it. We found that collective detection in mixed flocks of emberizid sparrows is limited markedly by relatively small degrees of visual and spatial separation between foragers. These limits on collective detection appear to influence the degree to which flock members lower their vigilance with increasing group size (the group size effect). Specifically, the decrease in collective detection with increasing visual and spatial isolation between foragers is accompanied by a concomitant decrease in the strength of the vigilance group size effect. Explanations for the vigilance-related effects of such separation based upon a bird’s ability to monitor the vigilance behavior of flockmates can be ruled out for our experimental system. Our results also shed light on the issue of whether the vigilance group size effect is influenced more by collective detection or the simple dilution of risk with increasing group size. We argue that collective detection is not only an important determinant of the group size effect, but also that the phenomena of collective detection and risk dilution are interdependent. Received: 25 July 1995/Accepted after revision: 17 December 1995     

机械故障振动信号征兆提取方法

在机械状态监测与故障诊断中，故障征兆的提取是关键环节，本文将介绍几种效果较好的故障征兆提取方法。