浅谈视频监控在石化安全领域的运用与探索

以天然气净化厂安全视频监控运用为实例,充分阐述了视频监控的分类、特点以及运用实效,对各类风险场景视频监控的应用配置进行了探讨,并针对问题和不足提出了措施建议,为视频监控在石化安全领域中的开发、运用、管理提供参考。     

基于Logistic曲线的城市应急能力评价研究

为科学、合理的对城市应急能力发展现状与发展过程进行评价,运用Logistic曲线模拟替代传统线性曲线分析城市应急能力发展过程,并运用该曲线对传统的线性隶属函数进行改进,提出了基于Logistic曲线的城市应急能力发展现状评价模型和城市应急能力可持续发展过程.运用该模型对广州市火灾应急能力进行实证分析,分析结果显示,20...     

直升机医疗救援的安全风险评价模型研究

为对直升机医疗救援任务实施有效的风险管理,提升其飞行安全水平,运用SHELL模型对直升机医疗救援飞行进行风险识别,确定了直升机医疗救援的风险因素.运用Borda序值法和风险矩阵法建立了 Borda-风险矩阵评价模型.运用Borda-风险矩阵评价模型对G通航进行实证分析,结果表明该评价模型具有可行性,可为直升机医疗救援风险评价提供参考.     

尾矿坝浸润线数据挖掘预测模型的样本选取研究

本文分别应用时间序列功能模型和回归模型,在原始数据的基础上建立样本,并运用支持向量回归机算法对样本进行训练,得出了尾矿坝浸润线埋深预测模型并进行了实例应用。研究证明,运用时间序列模型选取训练样本能够得出更为精确的预测结果。     

OHSMS绩效评价指标权重确定的研究

为了提高企业的职业健康安全管理体系绩效,有必要对其进行评价,构建评价指标体系并确定指标权重.运用层次分析法确定主观权重,运用灰色关联分析法确定客观权重,再运用乘法集成法得到组合权重,将组合权重进行排序,可以分析出对目标影响较大的指标.研究表明,将层次分析法和灰色关联分析法综合得到的组合权重,不仅考虑了专家的判断,也分析运用了实际数据,有效避免主观失误,提高了OHSMS绩效评价指标权重的科学性和可信度;权重的排序结果,显示了重要的评价指标,为实际工作提供了理论依据.     

混凝土强度检测在井筒支护中的应用

运用无损回弹测试方法，对井筒支护的强度进行现场检测，并依据测试结果运用数值分析的方法，对井筒安全稳定性进行分析，采取了加固补强的具体措施保证了东主井在一定时期内的安全生产。     

马尔科夫链模型在铁路春运客流预测中的应用

预测春运客流量是铁路部分的一项重要工作。运用马尔科夫链模型对春运期间一天中的客流量进行预测。首先,介绍了马尔科夫链模型及其预测的基本原理;其次,分析北京西站2007年春运期间每天的客流量,并按照北京西站突发事件三级预警方案将客流量数据处理为三个状态;最后,运用马尔科夫链模型对2007年的春运客流进行预测,结果表明,运用马尔科夫链模型具有良好的预测结果。     

用于桥式起重机挤压事故故障树分析的计算机软件

运用故障树分析的软件CAFTA,对桥式起重机挤压事故进行了故障树分析,探讨了事故发生的因果关系。     

基于 GIS 与 RS 的土壤重金属分布特征评价倡以武汉市为例

基于土壤样品中金属含量及空间位置数据,运用空间分析,对土壤重金属聚集特征进行分析,并结合遥感影像对重金属的分布及污染进行评估,具有极高的现实意义。以武汉为研究载体,运用土壤重金属分布评价方法进行研究,与研究区域实际状况基本相符,验证了方法的正确性和可操作性。     

神经元网络在安全评价中的应用

利用人工神经元网络进行安全评价可以克服传统的安全评价方法的缺点,提高安全评价方法的精确度和可靠性。笔者在BP神经网络基本原理的基础上,利用机会约束的思想建立了综合安全评价模型;运用反向传播算法和遗传算法对神经元网络进行训练;进而就系统综合安全评价模型进行求解,并对运用神经元网络进行综合安全评价的优点进行了分析。最后,通过对实例进行综合安全评价,得出计算结果,同时也证明将人工神经元网络应用于安全评价的可行性     

驾驶行为模型的研究进展

驾驶员行为模型的研究对于预测和干预驾驶员的风险行为、设计相关的道路安全设施与车内设备,以及制定交通法律法规等具有重要的意义。为了解和掌握学术界关于驾驶行为模型的研究进展,搜集、筛选和归纳了1960—2010年被SCI数据库索引的相关文章,将驾驶行为模型分类为描述性模型、信息处理模型、动机模型、计划行为理论(TPB)和躯体标识假设,并对每种模型进行评述和总结,理清这些模型间的内在联系。研究发现,现有各模型只是从某个角度研究驾驶员行为的部分特征,而不能解释驾驶员的全部行为。今后应不断完善和整合各类模型,并借鉴心理学、生理学和行为科学等相关领域的理论、知识,使驾驶行为模型变得更为实用、有效。     

Give them what they want or give them what they need? Ideology in the study of leadership

In recent years, a number of new, values‐based, or ideological models focusing on leader behavior have been proposed. These models include authentic, servant, character‐based, ethical, spiritual, and aesthetic leadership. In the present effort, we argue that these models, despite some differences in key dimensions, are tied together by a focus on moral behavior. The available evidence indicates that these models have only modest predictive power with respect to organizational performance criteria. More centrally, we argue that tests of these models are characterized by significant methodological problems with respect to both measurement and control. Moreover, these models suffer from some serious substantive concerns, including the explicit confounding of leadership and morality, discounting of system impacts, inappropriate assumptions about follower needs, and inappropriate scientific inferences. These models also fail to provide viable new approaches for leader development. We conclude that caution must be exercised when these models are employed as a basis for understanding leadership. Copyright © 2014 John Wiley & Sons, Ltd.     

林火数据的Logistic和零膨胀Poisson(ZIP)回归模型

利用统计模型对森林火灾数据进行了描述和分析,所用模型包括Logistic回归模型和零膨胀Poisson(ZIP)回归模型.将所用的森林火灾数据分别视为分组因子数据和有序变量数据进行建模.为了进行预测和验证,建模时使用部分数据,其余数据作为检验数据,用以检验预测的准确性.研究结果显示,所研究的两类模型都得到了与检验样本接近的结果,具有较好的火灾次数预测能力.其中零膨胀模型不仅可以得到与Logistic模型相当的结果,而且能够有效解决火灾数大于天数的问题,以及可以对零值过多的数据进行较好地建模.该研究表明所建立的Lo-gistic回归模型和零膨胀Poisson回归模型都是分析火灾与影响因素相关性的适用模型.     

10种事故致因模型的对比研究

为了给事故致因模型的研究提供理论支持,为日常管理中事故致因模型的选用提供参考依据,从事故的影响对象、模型的组成和事故发生的路径3个维度对10种事故致因模型进行了对比分析。研究结果表明:事故致因模型对事故影响对象的涵盖越来越全面;事故致因模型的组成日趋全面,且趋于对模型中各模块给出清晰定义以能保证分析结果的准确性;故发生的路径描述由链式向系统网状发展;根据充分性和效率的需求对模块化模型和非模块化的模型进行选择。     

Prediction models for psychophysical lifting capabilities and the resulting physiological responses for work shifts of varied durations

This paper presents models for predicting estimates of maximum weights of lift acceptable to industrial workers for 8-hour work shifts. Prediction models are also presented for heart rate and oxygen uptake at the maximum acceptable weight of lift. Experimental data collected on 37 males and 37 females were used in developing and verifying these models. Each subject performed nine variations of a lifting task involving three heights, four frequencies, and three box sizes to determine estimates of acceptable lifts for 8-hour shifts. The final lifting capability prediction models explained 76 to 82% variance in the experimental data. A procedure is also described to determine, from these models, acceptable lifts for work periods of variable lengths. The physiological-response-prediction models, which provide estimates of heart rate and oxygen uptake at the maximum acceptable weight of lift, were, however, relatively inferior to other similar response models available in the literature. This is thought to be due to the differences between the psychophysical and physiological methodologies.     

Experimental investigation of biodynamic human body models subjected to whole-body vibration during a vehicle ride

In this study, responses of biodynamic human body models to whole-body vibration during a vehicle ride were investigated. Accelerations were acquired from three different body parts, such as the head, upper torso and lower torso, of 10 seated passengers during a car ride while two different road conditions were considered. The same multipurpose vehicle was used during all experiments. Additionally, by two widely used biodynamic models in the literature, a set of simulations were run to obtain theoretical accelerations of the models and were compared with those obtained experimentally. To sustain a quantified comparison between experimental and theoretical approaches, the root mean square acceleration and acceleration spectral density were calculated. Time and frequency responses of the models demonstrated that neither of the models showed the best prediction performance of the human body behaviour in all cases, indicating that further models are required for better prediction of the human body responses.     

工业企业事故性泄漏扩散模型

对工业企业的泄漏扩散模型的国内外研究情况进行了调查研究,在对各模型优缺点分析的基础上,针对目前工业企业的事故后果模拟评价中所采用泄漏扩散模型存在的问题和不足,提出应用PG扩散模型研究物质泄漏的扩散模式,结合某企业的物质扩散算例说明该模型的实际应用,为工业企业的事故后果模拟评价以及重大事故应急预案的编制提供了一定的参考.     

Flamelet surface density modelling of turbulent deflagrating flames in vented explosions

The performance of two reaction rate models based on the laminar flamelet concept have been examined by calculating the behaviour of turbulent flame deflagration inside a semi-confined explosion tube. The models formulate the mean rate of reaction as a function of a transport equation for the flamelet surface density. The difference in the models is in modelling the source/sink terms of the flamelet surface density transport equation. The models are validated using laser diagnostics of flame deflagration in methane–air flammable mixture. The predictions are compared with experimental results for propagation, pressure history and flame speed. Sensitivity to cross-flow effects are investigated through comparison between two- and three-dimensional calculations. The numerically simulated results show that experimental trends are well reproduced by both models.     

The effect of overtaking behavior on unidirectional pedestrian flow

We present a model of overtaking behavior that can be used to simulate unidirectional pedestrian flow in routine. All pedestrians have the ability to determine whether or not to overtake other pedestrians according to their desired velocity and position. Although existing models such as cellular automata models, lattice gas models, social force models, etc., can be used to predict evacuation performance, most of these models are either computationally inefficient or do not account for some crucial elements of human behavior in a moving crowd. Furthermore, these models use either empirical equations developed from experiments or mechanical system analogies to determine movement decisions. The pedestrian flow patterns simulated by these models may deviate significantly from reality. In reality, pedestrians walk at different velocities and pedestrians with a higher walking velocity are accustomed to overtaking other pedestrians with a lower walking velocity and this paper aims to mimic this behavior as the original social force model developed by Helbing et al. does not reflect this pattern of collective pedestrian behavior. In this paper, we propose modifications of the social force model that reflects how overtaking behavior operates in routine. The comparison of the pedestrian flow pattern between the original social force model and the modified social force models with the real data collected by the camcorder is also performed in order to demonstrate our modified social force model can be used to achieve reasonable simulations of overtaking behavior among pedestrians.     

Revised fire consequence models for offshore quantitative risk assessment

Offshore oil and gas platforms are well known for their compact geometry, high degree of congestion, limited ventilation and difficult escape routes. A small mishap under such conditions can quickly escalate into a catastrophe. Among all the accidental process-related events occurring offshore, fire is the most frequently reported. It is, therefore, necessary to study the behavior of fires and quantity the hazards posed by them in order to complete a detailed quantitative risk assessment. While there are many consequence models available to predict fire hazards-varying from point source models to highly complex computational fluid dynamic models—only a few have been validated for the unique conditions found offshore.

In this paper, we have considered fire consequence modeling as a suite of sub-models such as individual fire models, radiation model, overpressure model, smoke and toxicity models and human impact models. This comprehensive suite of models was then revised by making the following modifications: (i) fire models: existing fire models have been reviewed and the ones most suitable for offshore conditions were selected; (ii) overpressure impact model: a model has been developed to quantify the overpressure effects from fires to investigate the possible damage from the hot combustion gases released in highly confined compartments; (iii) radiation model: instead of a point/area model, a multipoint grid-based model has been adopted for better modeling and analysis of radiation heat flux consequences. A comparison of the performance of the revised models with the ones used in a commercial software package for offshore risk assessment was also carried out and is discussed in the paper.