基于人工神经网络理论的建筑物火灾安全评价研究

依据建筑物火灾危险性的影响因素,应用人工神经网络理论及系统安全方法,建立了建筑物火灾危险性的评价指标体系,该方法摆脱了评价过程中的随机性和参评人员主观上的不确定性及其认识上的模糊性等缺点,大大提高了准确性。为了验证评价模型的准确性,将该理论应用到某高校图书馆火灾危险性评价中,快速、准确地得到了安全评价结果,取得了满意效果,为建筑物防火设计以及安全管理提供了可行的依据。     

航空维修人为差错影响因素分析中的模糊层次分析法

航空维修中的人为差错影响因素分析对于预防事故发生至关重要,如何定量分析及辨识出主要影响因素是亟待解决的问题。为此,结合航空维修实际,提出运用群组模糊层次分析法对人为差错影响因素进行量化排序和分类,归纳细化了影响因素层次体系,并给出了分析计算流程。以一起由维修人为差错所引发的航空事故为例进行了实证分析,结果表明:该方法能够辨识出航空维修中人为差错产生的主要影响因素,进而对制订最优事故预防方案,控制和减少该类事故的发生有积极意义。同时该方法对其他行业中的人为差错主要影响因素辨识有着一定的参考价值。     

Rethinking receiver operating characteristic analysis applications in ecological niche modeling

The area under the curve (AUC) of the receiver operating characteristic (ROC) has become a dominant tool in evaluating the accuracy of models predicting distributions of species. ROC has the advantage of being threshold-independent, and as such does not require decisions regarding thresholds of what constitutes a prediction of presence versus a prediction of absence. However, we show that, comparing two ROCs, using the AUC systematically undervalues models that do not provide predictions across the entire spectrum of proportional areas in the study area. Current ROC approaches in ecological niche modeling applications are also inappropriate because the two error components are weighted equally. We recommend a modification of ROC that remedies these problems, using partial-area ROC approaches to provide a firmer foundation for evaluation of predictions from ecological niche models. A worked example demonstrates that models that are evaluated favorably by traditional ROC AUCs are not necessarily the best when niche modeling considerations are incorporated into the design of the test.     

基于改进的SPAR-H化工实验室人因可靠性分析

近年来化工实验室事故屡禁不止,其中人的不安全行为是导致事故发生的主要原因。为评估实验人员可靠性,进一步管控化工实验室人的不安全行为,基于标准化工厂人因可靠性分析(SPAR-H)方法,结合化工实验室人因失误的特点,确定了更加适用于化工实验室的人因可靠性分析方法。首先依据“S-O-P”认知模型对人因失误类型进行划分;然后基于SPAR-H基本定义并结合化工实验室安全管理相关要求明确了性能影响因素(PSFs)水平确定标准;最后应用解释结构模型(ISM)确定不同失误类型对应的影响因素,并计算得到化工实验室人员实验过程中人因失误概率。结果表明:改进的SPAR-H方法评估流程与化工实验室人员实验操作情景相吻合,不仅能够量化人因失误的可能性,而且能够为化工实验室人的不安全行为管控制定针对性措施。     

Uncertainty Models for Estimates of Physical Characteristics of River Segments Over Large Areas

Spatially comprehensive estimates of the physical characteristics of river segments over large areas are required in many large‐scale analyses of river systems and for the management of multiple basins. Remote sensing and modeling are often used to estimate river characteristics over large areas, but the uncertainties associated with these estimates and their dependence on the physical characteristics of the segments and their catchments are seldom quantified. Using test data with varying degrees of independence, we derived analytical models of the uncertainty associated with estimates of upstream catchment area (CA), segment slope, and mean annual discharge for all river segments of a digital representation of the hydrographic network of France. Although there were strong relationships between our test data and estimates at the scale of France, there were also large relative local uncertainties, which varied with the physical characteristics of the segments and their catchments. Discharge and CA were relatively uncertain where discharge was low and catchments were small. Discharge uncertainty also increased in catchments with large rainfall events and low minimum temperature. The uncertainty of segment slope was strongly related to segment length. Our uncertainty models were consistent across large regions of France, suggesting some degree of generality. Their analytical formulation should facilitate their use in large‐scale ecological studies and simulation models.     

基于位置误差概率模型的物流无人机安全间隔评估方法研究*

针对未来多架物流无人机飞行在同一航线可能引起的安全冲突问题,提出规划无人机之间的安全间隔以规避运行风险。根据无人机在地面控制系统监控下的飞行特点,结合位置误差概率模型,综合考虑无人机定位误差、速度误差以及侧风的影响,依次建立无人机之间纵向、侧向和垂直方向的碰撞风险评估模型,并得出ICAO安全目标水平下的最小安全间隔。选取顺丰在江西省赣州市开辟的物流配送航线进行仿真,分析不同机型组合之间3个方向的最小安全间隔。研究结果表明:计算结果可以为物流无人机飞行配备安全间隔提供参考。     

基于层次分析法的Pedersen人因失误模型研究

科技的发展,使机器设备可靠性有了显著提高,工作环境也得到了改善,但由于人的复杂性,人因失误成为主要的致因因素,如何预防和控制人的失误就成为一个非常重要的课题.运用层次分析法的基本原理,结合Pedersen人因失误模型,构建了导致人因失误的层次分析模型,计算结果显示,人在某种状态下承受负荷的能力大小是影响其失误的主要原因.通过实例,证明了计算结果的可靠性,并提出了减少人因失误的建议.     

Pitfalls in molecular diagnosis in a family with severe factor VII (FVII) deficiency—misdiagnosis by direct sequence analysis using a PCR product

Molecular diagnostic tests are becoming a routine analysis in many laboratories. These modern analyses are widely used in clinical medicine, forensic, genetic and prenatal diagnosis and also in preimplantation genetic diagnosis. The accuracy of analysis is highly dependent on the success achieved in minimising genotyping errors. The pitfalls in molecular diagnostic tests can be due to a simple technique such as the polymerase chain reaction (PCR) used universally. This technique is routinely used for its apparent accuracy, but it is also a well-known source of errors. We report an error introduced during PCR reaction that leads to a wrong sequence result and consequently to a ‘false’ molecular result in a next prenatal diagnosis in a family with severe factor VII (FVII) deficiency. This error was verified using an unsuitable primer design in a rich repetitive sequence of the FVII gene that leads to a false annealing and then to a wrong molecular diagnosis. It is essential to link closely molecular data with clinical and phenotype analysis in order to avoid false-negative or false-positive results, which is of great importance to diagnosis and molecular prevention. Copyright © 2003 John Wiley & Sons, Ltd.     

反舰导弹射击精度评估研究

分析了反舰导弹射击精度评估领域命中概率与圆概率偏差、命中区域圆概率偏差的分歧,提出了制导概率椭圆偏差和制导概率圆偏差的新概念。将反舰导弹飞行末段雷达导引头波束覆盖区域与打击目标水面舰艇轮廓大小进行了比对,根据比对结果划分为雷达导引头波束完全覆盖、部分覆盖和内嵌于目标三个等级,对应提出了命中概率评估方法、制导概率椭圆偏差评估方法和制导概率圆偏差评估方法。通过反舰导弹对快艇、驱护舰和航母三类目标射击精度评估算例验证了反舰导弹射击精度评估总体方案的全面性和三类方法的适用性。     

研究星载IPDA激光雷达探测CO2光谱纯度

光谱纯度是积分路径差分吸收激光雷达最重要的系统参数之一,光谱纯度直接影响CO₂数据的探测精度.在模式研究中要求输入观测的CO₂数据误差小于1×10^-6,对激光器光谱纯度参数设计提出很高要求.本文采用有效吸收截面分析探测反演误差的方法,研究了由光谱纯度带来的CO₂柱浓度探测误差,并基于窄带滤波器对光谱不纯能量的抑制作用,对不同带宽的窄带滤波器进行选择和分析,从而达到降低对光谱纯度的要求,提升探测精度的目的,最后讨论了由于窄带滤波器造成的能量衰减所对随机误差的影响.结果表明：当光谱纯度为99.9%,窄带滤波器带宽1GHz,透射率为0.86,脉冲能量为100mJ时CO₂探测的系统误差小于0.084×10^-6,随机误差小于0.02×10^-6,满足探测精度要求.