基于高斯过程机器学习算法的矿柱稳定性分析

矿柱是地下矿山支撑顶板围岩、维持采场稳定的关键结构要素。为迅速准确地判别矿柱稳定性,选取矿柱宽度、矿柱高度、矿柱宽高比、矿岩单轴抗压强度和矿柱承受载荷作为影响指标,利用高斯过程机器学习算法建立矿柱状态与其主要影响因素之间的非线性映射关系,进而提出一种基于高斯过程二元分类(GPC)的矿柱状态识别模型。结合工程实例,以40组样本数据进行训练,以10组样本数据对该模型进行检验,并与ANN和SVM进行对比。结果表明,矿柱状态识别的高斯过程模型是科学可行的,该模型具有参数自适应化获取、分类精度高、计算复杂度低等优点,还可对矿柱状态判别结果的不确定性或可信度进行定量化评价。     

Experimental study on leak detection and location for gas pipeline based on acoustic method

The leak of gas pipelines can be detected and located by the acoustic method. The technologies of recognizing and extracting wave characteristics are summarized in details in this paper, which is to distinguish leaking and disturbing signals from time and frequency domain. A high-pressure and long distance leak test loop is designed and established by similarity analysis with field transmission pipelines. The acoustic signals collected by sensors are de-noised by wavelet transform to eliminate the background noises, and time-frequency analysis is used to analyze the characteristics of frequency domain. The conclusion can be drawn that most acoustic signals are concentrated on the ranges of 0-100 Hz. The acoustic signal recognition and extraction methods are verified and compared with others and it proves that the disturbing signals can be efficiently removed by the analysis of time and frequency domain, while the new characteristics of the accumulative value difference, mean value difference and peak value difference of signals in adjacent intervals can detect the leak effectively and decrease the false alarm rate significantly. The formula for leak location is modified with consideration of the influences of temperature and pressure. The positioning accuracy can be significantly improved with relative error between 0.01% and 1.37%.     

基于面部特征识别的管制员疲劳监测方法研究

为进一步控制疲劳诱发的空管人为差错,提出一套基于面部特征识别的空中交通管制员疲劳实时监测方法。应用眼动仪和视频记录系统开展36 h睡眠剥夺试验,确定PERCLOS值、平均闭眼时长、哈欠频率3个疲劳判定指标的判定阈值,并给出综合疲劳程度的融合算法。应用机器视觉的理论和方法,开发功能完整的管制员疲劳状态实时监测告警原型系统。结果表明,通过对面部特征的识别和状态融合能够实现对管制员疲劳状态的实时监测。建议在此基础上进一步开发可实用的疲劳监测系统。     

矿井出入口智能检控系统的设计与实现

本文在分析国内煤矿井口安全监控系统的技术现状和矿井安全生产需求的基础上,结合四川中小煤矿企业的井口安全管理需要,提出了一种集人员考勤和危险源管理功能于一体的矿井出入口检控系统的设计方案。系统将活体指纹识别、视频技术与井口检控有机结合起来,实现出入井人员活体指纹考勤功能、对井下各地点安全隐患排查数据录入、分类整理、督促整改、隐患预警、查询、隐患打印、隐患复查等隐患治理控制功能,从根本上解决小煤矿管理不到位、技术人员缺乏、工作效率低等问题。加强了矿井安全管理力度,为安全隐患排查治理工作的规范、科学、高效提供了强有力保障。     

福州第二水源山仔水库营养类型的判定及变化趋势分析

对1998～2002年山仔水库监测数据．运用属性识别方法进行评价分析,其结果表明,这5年来湖泊富营养化程度有明显减弱的趋势,但1998年水体达极富营养状态．在2001年后水质有明显好转。通过判定该水库5年的营养状态,分析其变化趋势．可找出引起富营养化的原因。     

应用时间-活动模式估计儿童个体NOx暴露水平

将时间-活动模式与微环境空气污染物监测相结合估计儿童个体NO_x暴露水平. 分别在北京和青岛两市各选择3所学校监测校园环境污染物,并在每个学校选择10名儿童监测其家庭居室空气中ρ(NO_x)及个体24 h NO_x暴露水平,同时收集其时间-活动信息,采用时间-活动模式估计儿童的个体暴露水平. 交通工具中的NO_x暴露水平可通过校门口空气中ρ(NO_x)和交通工具的污染水平系数(λ)估计.结果表明:利用时间-活动模式模拟的儿童个体24 h NO_x暴露水平为0.041 mg/m3,与调整后的个体暴露实测水平相关性较好(R0.785,P<0.01),差值为(-0.002　6±0.013　0) mg/m3,二者的差异无统计学意义 (t0.139,P>0.05).表明采用时间-活动模式与微环境空气质量监测结果相结合方法估计的儿童NO_x暴露水平与个体实际暴露水平一致.      

The interplay between genetic and environmental effects on colony insularity in the clonal invasive little fire ant <Emphasis Type="Italic">Wasmannia auropunctata</Emphasis>

The little fire ant, Wasmannia auropunctata, constitutes one clonal supercolony throughout Israel, providing an opportunity to examine the effects of genotype versus environment on nestmate recognition. Intraspecific encounters among field-collected or among laboratory-maintained colonies were nonaggressive, but encounters between freshly collected and laboratory-maintained colonies were highly aggressive. Analyses of cuticular hydrocarbons revealed that freshly field-collected colonies had distinguishable profiles. Moreover, freshly collected colonies had profiles disparate from those of the same colonies after 4 months in the laboratory. These results indicate a strong interplay between genetic-based and environmentally based effects on the recognition cues. We propose that in the field the ants’ diet breadth is broad and consequently the incorporation of diet-borne substances is insufficient to mask the genetically determined cues. In the laboratory, however, the restricted diet promoted the incorporation of alien hydrocarbons at high levels, thus altering the genetically based cues to the point of alienation. These results shed a new light on the mechanisms by which environmental cues may affect label and/or template formation in ants. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

基于属性识别理论的城市生态安全评价——以广州市为例

城市生态安全评价可以使管理者和公众及时了解城市发展中生态环境压力的严重态势,为城市可持续发展战略的制定和生态系统的管理提供依据和指导。文章构建了由活力、组织结构、恢复力、生态系统服务功能和人类健康状况5个要素构成的城市生态安全评价指标体系。在有序分割类和属性识别准则的基础上,建立了城市生态安全评价的属性识别模型。采用层次分析法（AHP）和熵权法赋予指标权重,充分利用了主观和客观评价方法的优势。根据此模型对广州市1991—2007年间的生态安全状况进行了实证分析。结果表明城市生态安全状况呈现出,从＂较不安全级＂上升到＂临界安全＂级。在5个评价要素中,除人群健康状况外,其他4个评价要素呈现出逐年改善的趋势。属性识别理论能对事物进行有效识别和比较分析,且原理直观、计算简便、评价结果准确,在评价、预测、决策等领域中具有广泛的适用性。     

Queen acceptance and the complexity of nestmate discrimination in the Argentine ant

In most social insect species, individuals recognize and behave aggressively towards non-nestmate conspecifics to maintain colony integrity. However, introduced populations of the invasive Argentine ant, Linepithema humile, exhibit pronounced variation in intraspecific aggression denoting diversity in nestmate recognition behavior, which possibly shapes their social structure and the varying levels of unicoloniality observed among these populations. One approach to better understand differential aggression behaviors towards conspecifics and recognition cue perception and response in L. humile is to examine variation in nestmate discrimination capability among genetically distinct colonies under different social contexts. Consequently, we investigated the dynamics of queen and worker recognition in southeastern US L. humile queenless and queenright colonies by measuring rates of non-nestmate worker and queen adoption and intercolony genetic similarity. Aggression levels between colony pairs differed and were associated with non-nestmate worker, but not queen adoption. Adoption of queens and workers was a function of host colony origin, while colony queen number affected adoption of queens, but not workers, with queens more readily accepted by queenless hosts. Fecundity of adopted non-nestmate queens was comparable to that of rejected non-nestmate and host colony queens, suggesting that queen fecundity did not affect adoption decisions. Genetic similarity between colonies ranged from 30 to 77% alleles shared, with more genetically similar colonies showing lower levels of intraspecific aggression. Non-nestmate queens and workers that were more genetically similar to host colony workers were more likely to be adopted. We provide the first evidence for the role of L. humile colony queen number on queen discrimination and suggest an effect of resident queens on worker conspecific acceptance thresholds. Our findings indicate a role for genetically based cues in L. humile nestmate recognition. However, subtle discrimination capability seems to be influenced by the social context, as demonstrated by more frequent recognition errors in queenless colonies.     

图像识别技术及其在机械零件无损检测中的应用

无损检测法是一种常用的故障诊断技术,故障诊断从本质上来讲就是模式识别问题,而模式识别又可以狭义地理解为图像识别。从介绍图像、图像识别、图像识别过程和图像识别系统的基本概念着手,就几种常用图像识别方法的原理和特点进行比较,给出了CCD图像获取系统的组成。最后,结合发动机曲轴的一种自动磁粉探伤系统实例,对系统的图像处理和识别流程进行详细的讨论,并针对一般无损检测系统难以满足曲轴的检测要求和精度要求的状况,提出经过改进的一种适用于曲轴的整体无损检测系统。该系统有助于高效和完整地获取整个曲轴的图像,提高图像信息的质量,从而提高发动机曲轴表面缺陷检测的准确性和可靠性。