Estimating dyad association probability under imperfect and heterogeneous detection

In animal behaviour studies, association indices estimate the proportion of time two individuals (i.e. a dyad) spend in association. In terms of dyads, all association indices can be interpreted as estimators of the probability that a dyad is associated. However, traditional indices rely on the assumptions that the probability to detect a particular individual (p) is either approximately one and/or homogeneous between associated and not associated individuals. Based on marked individuals we develop a likelihood based model to estimate the probability a dyad is associated (ψ) accounting for p < 1 and possibly varying between associated and not associated individuals. The proposed likelihood based model allows for both individual and dyadic missing observations. In addition, the model can easily be extended to incorporate covariate information for modeling p and ψ. A simulation study showed that the likelihood based model approach yield reasonably unbiased estimates, even for low and heterogeneous individual detection probabilities, while, in contrast, traditional indices showed moderate to strong biases. The application of the proposed approach is illustrated using a real data set collected from a population of Commerson's dolphin (Cephalorhynchus commersonii) in Patagonia Argentina. Finally, we discuss possible extensions of the proposed model and its applicability in animal behaviour and ecological studies.     

Knowledge gain and behavioral change in citizen-science programs

Citizen-science programs are often touted as useful for advancing conservation literacy, scientific knowledge, and increasing scientific-reasoning skills among the public. Guidelines for collaboration among scientists and the public are lacking and the extent to which these citizen-science initiatives change behavior is relatively unstudied. Over two years, we studied 82 participants in a three-day program that included education about non-native invasive plants and collection of data on the occurrence of those plants. Volunteers were given background knowledge about invasive plant ecology and trained on a specific protocol for collecting invasive plant data. They then collected data and later gathered as a group to analyze data and discuss responsible environmental behavior with respect to invasive plants. We tested whether participants without experience in plant identification and with little knowledge of invasive plants increased their knowledge of invasive species ecology, participation increased knowledge of scientific methods, and participation affected behavior. Knowledge of invasive plants increased on average 24%, but participation was insufficient to increase understanding of how scientific research is conducted. Participants reported increased ability to recognize invasive plants and increased awareness of effects of invasive plants on the environment, but this translated into little change in behavior regarding invasive plants. Potential conflicts between scientific goals, educational goals, and the motivation of participants must be considered during program design.     

基于Hilbert谱信息熵的煤矸放落振动特征分析

针对煤矿安全生产中的综放工作面煤矸界面探测问题,提出利用煤矸下落冲击钢板的振动特征来探测煤矸界面的方法。煤矸振动信号表现出非平稳特征,采用经验模态分解方法将复杂矿井环境下的煤矸振动信号分解成固有模态分量。选择包含煤矸振动特征的前7个本征模函数(IMF)分量,通过Hilbert变换得到Hilbert谱。分析不同放煤状态下钢板振动信号的Hilbert谱发现,顶煤下落时的Hilbert谱分布较均匀,而煤矸混放时的Hilbert谱呈现不均匀分布。根据信息熵理论,提出了基于Hilbert谱信息熵的煤矸振动特征提取方法。试验结果表明,顶煤下落时的Hilbert谱信息熵要大于煤矸混放时的Hilbert谱信息熵,因此,煤矸振动的Hilbert谱信息熵特征能够准确地反映放煤状态。     

基于无人机的大气细颗粒物与臭氧污染探测与溯源研究进展

近年来,中国大气细颗粒物(PM_2.5)污染的治理已取得阶段性成效,但臭氧(O₃)污染快速上升,实现PM_2.5和O₃协同控制的基础与关键是针对大气污染物的精细化探测和污染溯源.随着无人机技术和传感器技术的迅速发展,基于无人机平台的大气污染探测可以有效获得近地层的PM_2.5和O₃结构特征,并结合计算机算法对大气污染事件进行精准溯源,具有高时效性、高灵活性和高时空分辨率的特征,有助于研究人员了解区域污染物的分布、变化以及来源,为大气复合污染的协同控制提供科学依据.通过回顾传统的大气污染探测方法,总结了污染探测领域常用的无人机飞行平台类型和探测仪器,归纳了基于无人机的PM_2.5和O₃污染探测应用与相关溯源算法,并展望了无人机大气探测的未来研究方向.     

极板转运装置自动控制系统的设计与实现

介绍了极板转运装置自动控制系统的设计思想.详细阐述了系统的组成,检测开关的现场布置以及系统程序的设计.     

Deriving heart rate variability indices from cardiac monitoring—An indicator of driver sleepiness

Objective: Driver fatigue is considered to be a major contributor to road traffic crashes. Cardiac monitoring and heart rate variability (HRV) analysis is a candidate method for early and accurate detection of driver sleepiness. This study has 2 objectives: to evaluate the (1) suitability of different preprocessing strategies for detecting and removing outlier heartbeats and spectral transformation of HRV signals and their impact of driver sleepiness assessment and (2) relation between common HRV indices and subjective sleepiness reported by a large number of drivers in real driving situations, for the first time.

Methods: The study analyzed >3,500 5-min driving epochs from 76 drivers on a public motorway in Sweden. The electrocardiograph (ECG) data were recorded in 3 studies designed to evaluate the physiological differences between awake and sleepy drivers. The drivers reported their perceived level of sleepiness according to the Karolinska Sleepiness Scale (KSS) every 5?min. Two standard methods were used for identifying outlier heartbeats: (1) percentage change (PC), where outliers were defined as interbeat intervals deviating >30% from the mean of the four previous intervals and (2) standard deviation (SD), where outliers were defined as interbeat interval deviating >4 SD from the mean interval duration in the current epoch. Three standard methods were used for spectral transformation, which is needed for deriving HRV indices in the frequency domain: (1) Fourier transform; (2) autoregressive model; and (3) Lomb-Scargle periodogram. Different preprocessing strategies were compared regarding their impact on derivation of common HRV indices and their relation to KSS data distribution, using box plots and statistical tests such as analysis of variance (ANOVA) and Student’s t test.

Results: The ability of HRV indices to discriminate between alert and sleepy drivers does not differ significantly depending on which outlier detection and spectral transformation methods are used. As expected, with increasing sleepiness, the heart rate decreased, whereas heart rate variability overall increased. Furthermore, HRV parameters representing the parasympathetic branch of the autonomous nervous system increased. An unexpected finding was that parameters representing the sympathetic branch of the autonomous nervous system also increased with increasing KSS level. We hypothesize that this increment was due to stress induced by trying to avoid an incident, because the drivers were in real driving situations.

Conclusions: The association of HRV indices to KSS did not depend on the preprocessing strategy. No preprocessing method showed superiority for HRV association to driver sleepiness. This was also true for combinations of methods for frequency domain HRV indices. The results prove clear relationships between HRV indices and perceived sleepiness. Thus, HRV analysis shows promise for driver sleepiness detection.     

A new fault detection method for non-Gaussian process based on robust independent component analysis

Conventional fault detection method based on fast independent component analysis (FastICA) is sensitive to outliers in the modeling data and thus may perform poorly under the adverse effects of outliers. To solve such problem, a new fault detection method for non-Gaussian process based on robust independent component analysis (RobustICA) is proposed in this paper. A RobustICA algorithm which can effectively reduce the effects of outliers is firstly developed to estimate the mixing matrix and extract non-Gaussian feature called independent components (ICs) by robust whitening and robust determination of the maximum non-Gaussian directions. Furthermore, a monitoring statistic for each extracted IC is constructed to detect process faults. Simulations on a simple example of the mixing matrix estimation and a fault detection example in the continuous stirred tank reactor system demonstrate that the RobustICA achieves much higher estimation accuracy for the mixing matrix and the ICs than the commonly used FastICA algorithm, and the RobustICA-based fault detection method outperforms the conventional FastICA-based fault detection method in terms of the fault detection time and fault detection rate.     

PVC电缆火灾早期特征气体的组成分析和传感器探测

利用气体传感器对电缆绝缘过热释放的气体进行监测有望实现电气火灾早期预警,目前该技术的发展由于特征气体未得到确认而受到限制。分析了聚氯乙烯(PVC)电缆绝缘层在不同温度下释放的气体组成,并对其主要成分进行传感器测试。TG-IR和GC-MS分析表明PVC电缆绝缘层在200℃前已出现微小失重,释放出以邻苯二甲酸二辛酯(DOP)等塑化剂为主的气体。气敏测试中,商用半导体气体传感器对DOP和电缆蒸气产生较好、相似的响应性能。因此,DOP可以作为PVC电缆火灾早期的特征气体。     

基于U-Net深度神经网络的早期火灾烟雾自动分割方法

在火灾事件监测中,为了减少数据处理量、加快探测速度,需要先分割出疑似烟雾区域。传统的烟雾分割算法大多需要设置阈值进行处理,算法的环境适应能力还需进一步提升。在研究中,使用U-Net结构的深度神经网络进行早期火灾烟雾的自动分割,通过半自动算法人工辅助分割出烟雾区域的图像样本,基于深度神经网络对分割烟雾区域进行学习,得到原始视频帧到分割结果的映射模型,并据此模型进行烟雾区域分割。在测试集上的分割实验结果表明该方法与传统方法相比,不需要设置阈值,自动化程度更高,分割速度极快,在疑似烟雾区域分割任务中性能较好。     

基于BP神经网络的受限空间火灾联合探测方法

针对航空受限空间火灾探测高误报的问题,在现有技术成果基础上对多种火灾探测方式进行研讨,并提出1种基于BP神经网络技术的飞机机身内部受限空间火灾联合探测报警方法。该方法结合现有烟雾感应、气体传感器探测等常用火灾探测技术,以红外热成像探测为辅助手段,采用神经网络实现数据融合,对模拟实验舱火灾烟雾进行联合探测,在单一火灾探测方式基础上提高了探测准确率。