NaI（Tl）多道谱仪测量固体样品时实验数据准确性分析

通过对同一样品中的放射性核素含量进行多次等精度测量,对结果的可靠性进行分析和出现离散的数据进行数据舍弃检验。在放射性测量过程中除考虑统计误差外,同时也应考虑仪器的、人为的及偶然因素导致的系统误差,从而提高测量精度。     

Accuracy assessment of a small-area method for estimating the spatial distribution of the degree of tree damage

Tree damage, gauged by the amount of defoliation, is one of the basic criteria used to determine treatments for protected and economic forests. Monitoring should include an assessment of the degree of tree damage in different spatial scales. Therefore, in addition to the commonly applied large-area methods, small-area methods should be used. The aim of the paper is to present the results of the accuracy assessment of a small-area method, proposed by Podlaski (2005) [Podlaski, R. (2005). Inventory of the degree of tree defoliation in small areas. Forest Ecology and Management, 215, 361–377], for monitoring the degree of tree damage. The degree of tree damage was shown in sub-blocks P₃ of the system of information on natural environment (SINUS). To estimate the spatial distribution of the degree of tree defoliation, survey sampling, based on simple random sampling with replacement (SRSWR), was used. The degree of damage to fir (Abies alba Mill.) and beech (Fagus sylvatica L.) was analysed in the Święty Krzyż forest section in the Świętokrzyski National Park. The maximum total estimation errors for the proportion of trees with a degree zero of damage, and with second and third degrees of damage together (for α = 0.05) were at most 30.8% for fir and 24.3% for beech trees. For standard, small-area evaluations, these are satisfactory values. In the Święty Krzyż forest section, the number of P₃ sub-blocks with 0.00–5.00% of undamaged trees and with 80.01–100.00% of moderately- or severely-damaged trees was significantly greater for fir than for beech. These results indicate that the fir population was unhealthier than the beech group in the study area. P₃ sub-blocks of the SINUS system, in which the proportion of the healthiest trees was highest, were situated at the forest margin, bordering on meadows and arable fields (in the case of fir) and forming dense patches consisting of several sub-blocks, or occurring singly in the whole study area (in the case of beech). The results show the significant differentiation of forest tree health in small areas.     

Method for identifying errors in chemical process development and design base on accidents knowledge

It has been claimed that the high accident rate in the chemical process industry is due to poor dissemination of accident knowledge that affects directly the level of learning from accidents. In response to this situation, this paper utilized past accident knowledge as a basis to develop a safety oriented design tool whereby the accident information were directly disseminated into plant design. The method was developed based on our previous accident analysis of design error in which the common design errors were ranked in accordance to their frequency and its origins during normal plant design project. Based on the design error ranking and its origin at a specific design phases, a method for design error detection is proposed. The method is expected to be able to identify the possible design error and its causes throughout chemical process development and design. The main objective is to trigger safe design thinking at the specific design phases so that appropriate action for risk reduction could be timely implemented. The Bhopal and BP Texas tragedies are used as case studies to test and verify the method. The proposed method can detect up to 74% of design errors.     

New considerations for SIL verification of functional safety fieldbus communication

Safety integrity level (SIL) verification of functional safety fieldbus communication is an essential part of SIL verification of safety instrumented system (SIS), and it requires quantifying residual error probability (RP) and residual error rate of function safety communication. The present quantification method of residual error rate uses RP of cyclic redundancy check (CRC) to approximately replace the total RP of functional safety communication. Since CRC only detects data integrity-related errors and CRC has intrinsically undetected error, some other residual errors are not being considered. This research found some residual errors of the present quantification method. Then, this research presents an extended new approach, which takes the found residual errors into account to determine more comprehensive and reasonable RP and residual error rate. From perspective of the composition of safety message, this research studies RPs of those controlling segments (sequence number, time expectation, etc.) to cover the found residual errors beyond CRC detection coverage, and the influences of insertion/masquerade errors and time window on RP are investigated. The results turn out these residual errors, especially insertion/masquerade errors, may have a great influence on quantification of residual error rate and SIL verification of functional safety communication, and they should be treated seriously.     

基于改进的最小二乘法瓦斯涌出量预测

通过对基于绝对误差最小的传统最小二乘法与在此基础上考虑相对误差的改进最小二乘法的比较,得知改进的最小二乘法预测更为准确。现场应用于鹤壁四矿的相对瓦斯涌出量随深度变化的预测,建立了瓦斯涌出量预测模型,通过预测值与实际值比较,证明改进的最小二乘法预测鹤壁四矿瓦斯涌出量是可行的、有效的,对指导煤矿安全生产具有重要意义。     

基于信息处理和事故链原理的结构化飞行员差错模型

为了对民航飞行员人为差错进行系统分析,研究民航业中常见人为差错的分析方法,指出其局限性。基于信息处理过程和事故链原理,从微观的认知过程和宏观的HFACS框架2个层面构建一种新的飞行员人为差错分析模型,并对模型中的情景分析、差错模式与分类、认知机理、差错恢复、差错成因、反馈机制和防御机制等几大核心模块进行阐述,重点介绍认知机理在差错形成中产生的关键影响因素,提出结构化差错分析方法的思路与步骤。该模型为民航飞行员人为差错的产生机理、差错分类和差错成因的分析提供一种新思路,但模型的进一步应用还有待于通过案例分析来验证。     

管制员人为差错影响因素及指标权重分析

根据SHELL模型,将影响管制员人为差错的因素从内部因素和外部因素进行分析,内部因素包括管制员生理因素、心理因素、管制员业务技能三个部分;外部因素包括管制员与管制员班组之间、管制员与管制设备之间、管制员与管制程序等软件之间。通过分析和专家咨询,建立了影响管制员人为差错的影响因素指标体系,在模糊层次分析法的基础之上引入0.1-0.9标度法的三角模糊数,通过对各影响因素进行对比分析,得出了影响管制员人为差错的三级指标的重要度排序,排在前八位的因素依次为酒精及药物影响、形势意识、班组人员性格搭配、班组人员能力搭配、疲劳、人机界面、外界压力、情绪状态。为管制员培训和管理层进行安全检查提供了基础理论依据。     

基于实船结构监测数据的异常检测及处理方法

目的 获得高质量实船结构监测数据,减少数据因外界干扰造成的异常现象,现开发一套面向真实海况的实船结构监测数据的异常处理方法。方法 将统计学中的Z-score异常检测方法与数据处理中常用的平均值计算法相结合,实现对船舶结构应力数据的异常处理。基于正常信号创建含有异常现象的验证数据,对新的异常处理方法进行精度验证。结果 相比Hampel滤波法、Smooth平滑函数等传统的信号处理方法,Z-score异常检测及平均值计算方法的异常处理精度最高,且以此为基础计算的结构信号统计值准确度较好。结论 Z-score异常检测及平均值计算方法可实现真实海况下的实船结构监测数据的异常处理,并在结构应力数据的价值挖掘上可提供有力支撑。     

基于人类动力学的地震压埋人员先验在室率研究

破坏性地震发生后,科学、有效地评估地震压埋人员压埋情况,是有序地开展应急救援工作的基础。人员在室率作为压埋率评估模型的主要参数之一,目前主要以作息时间区为基础以省或市为基础单元进行计算,这难以体现基础单元之下的不同市或县人员在室率之间的差异性。为解决这一问题,采用人类动力学相关原理,运用极大似然法估算人员在室率,进而形成基于人类动力学的地震压埋人员压埋率预估模型。最后以汶川8.0级地震部分受灾学校为例,进行压埋率实际评估和模型误差分析。研究表明:提出的微观在室率能更准确地反映人员在室情况具有更好的适用性;在实际评估中模型精度达到±0.13,表明模型能达到较好的预估效果,能为震后应急救援工作提供一定的参考。     

ANP-Fuzzy方法在MPL学生威胁与差错管理能力评估中的应用

为考核CCAR141飞行学校多人制机组驾驶员执照(Multi-crew Pilot License,MPL)训练效果,对其核心单元——威胁与差错管理(Threat and Error Management,TEM)能力进行评估.在分析ANP原理和模糊综合评价方法的基础上,结合ICAO的PANS-TRG文件,建立了面向MPL培训的TEM能力评估指标体系和能够精确界定定性指标的Fuzzy多层次评价结构.采用ANP-Fuzzy方法对CCARI41飞行学校C的MPL学生TEM能力进行评估.结果表明,威胁管理比差错管理、非预期状态管理对MPL学生TEM能力的影响大,同时,该方法能够有效区分学生自身不同方面及不同学生间的管理能力差异.此外,借助Super Decisions软件也可用于其他训练单元的效果评估.