采用STAMP-24Model的多组织事故分析

安全生产事故往往由多组织交互、多因素耦合造成,事故原因涉及多个组织。为预防和遏制多组织生产安全事故的发生,基于系统理论事故建模与过程模型(Systems-Theory Accident Modeling and Process,STAMP)、24Model,构建一种用于多组织事故分析的方法,并以青岛石油爆炸事故为例进行事故原因分析。结果显示:STAMP-24Model可以分组织,分层次且有效、全面、详细地分析涉及多个组织的事故原因,探究多组织之间的交互关系;对事故进行动态演化分析,可得到各组织不安全动作耦合关系与形成的事故失效链及管控失效路径,进而为预防多组织事故提供思路和参考。

基于24Model-D-ISM的地铁站火灾疏散影响因素研究

为预防地铁站火灾事故,深入了解地铁站火灾人员疏散影响因素间的内在联系与层次结构,基于第6版“2-4”模型(24Model)分析63起地铁站火灾疏散事故,充分考虑各个因素之间的交互作用,提取19个影响地铁站人员疏散的关键因素,建立地铁站火灾人员疏散影响因素指标体系;采用算子客观赋权法(C-OWA)改进决策试验与评价实验法(DEMATEL),确定地铁站火灾人员疏散的重要影响因素;在此基础上,采用解释结构模型(ISM)分析各个因素间的层次结构及相互作用路径,构建地铁站火灾人员疏散影响因素的多级递阶结构模型。研究结果表明:疏散引导、恐慌从众行为、人员拥挤为地铁站火灾人员疏散的关键影响因素;地铁站火灾人员疏散受表层因素、中间层因素、深层因素共同作用的影响,其中,疏散教育与培训、设施维护与检查、疏散预案等因素是根源影响因素,重视根源影响因素的改善有利于从本质上预防和控制事故的发生。

基于改进24Model-ISM-SNA建筑工人不安全行为关联路径研究

建筑施工现场环境复杂,为有效控制不安全行为发生,基于行为安全“2-4”模型对360份具有代表性的建筑安全事故调查报告进行分析,提取出22个不安全行为的主要影响因素。利用灰色关联分析方法(GRA)改进的集成ISM-SNA模型,将不安全行为风险因素划分为表层、过渡层与深层,然后对风险因素进行可视化分析、中心度分析及凝聚子群分析,揭示了各致因因素间的关联关系和传导路径。结果表明,建筑工人不安全行为影响因素可划分成7级3阶的多级递阶结构,安全意识、现场监管、外部环境是建筑工人不安全行为的关键影响因素,同时现场监管和隐患排查到位能有效降低不安全行为的发生。

ChatModeler:基于大语言模型的人机协作迭代式需求获取和建模方法

需求获取和建模是需求工程中的关键步骤,影响后续系统设计与实现.传统的需求获取和建模方法通常由需求提供者、需求分析师等多类干系人共同协作、反复迭代完成,需要耗费大量的人力.如何减轻需求提供者与需求分析师的负担、提高获取和建模的效率有着重要意义.现有工作中有的使用知识库来提供更多知识,以辅助获取或者建模,有的利用自然语言处理等技术对获取或者建模过程进行自动化,但是它们并没有减轻需求提供者的负担.利用大语言模型(large language models,LLMs)的生成能力,提供了一种人机协作的迭代式需求获取和建模框架ChatModeler.具体来说,根据真实世界中需求团队的分工及协作关系,将部分需求提供者、需求分析师等角色的工作由大语言模型承担,而需求提供者只需要进行确认.为大语言模型扮演的各种角色进行了提示词设计,该提示词会随需求的元模型而变化.ChatModeler在7个需求案例上与3种需求模型的自动建模方法进行了14组对比实验,证明了ChatModeler在降低需求提供者的负担和生成高质量需求模型2个方面上的优越性.

24Model与LCM原因因素定义对比研究

为探究损失致因模型(LCM)原因因素定义与事故致因“2-4”模型(24Model)存在的异同和优缺点,梳理2个模型各层面原因和结果的定义,对比定义内容及其对事故原因分析等安全实务的指导作用,并以一起瓦斯爆炸事故为例加以实证分析,获得二者分析结果之间的差异。研究结果表明:LCM是首个将管理因素纳入事故致因分析的一维事件序列模型,可明确各层面原因因素的定义和因素间的逻辑关系,但部分定义存在交叉重复的问题,并没有揭示安全工作指导思想等深层次事故致因因素;24Model作为系统性事故致因模型,对各类因素的定义均以组织为主体,描述事件、事故、安全的概念内涵,划分个体安全动作、安全能力和组织安全管理体系的类别并给出含义解析,探究组织安全文化层面的问题并以32个元素体现;2个模型的事故原因分析方法均建立在对各层级原因因素定义的基础上,并适用于模型理论体系本身。

基于Modelica的飞行器多专业统一模型研究

结合以飞行器系统为代表的现代航天产品在研制过程中面临的的挑战,针对在总体设计阶段的验证与分析需求,采用多专业统一建模语言Modelica构建了飞行器各主要系统的模型,通过模型的集成仿真,实现了飞行器系统在总体设计阶段的验证。再基于所构建的飞行器模型,进行实时代码生成,进而实现了信号级实时仿真系统,并开展了快速控制原型系统仿真。以上工作为MBSE方法在飞行器系统设计过程中进行了探索与铺垫。

耦合优化蚁群算法与P-Median model的选址模型设计

为节省城建部门对于公共体育设施的投入成本以及提高城市人民生活质量,以运动场所优化选址为例,提出一种新型设施选址模型。该模型主要基于P-Median model(最小化阻抗模型)根据需求点数量从全部候选设施选址中选择设施空间位置,让用户达到离自己最近设施距离成本总和最小的目的,对选址的基本原则和实际情况提出要求,构造目标函数用于优化后蚁群算法求解进行选址工作。优化蚁群算法实现基于Python语言模块,通过改进蚁群原始信息素,提升原有算法的收敛速度,求出目标函数最优解,可以很好地模拟对于运动场所的选址。用二者耦合进行优势互补所设计的选址模型来搜寻研究区蚁群信息素浓度残留最大的栅格像元,从而确定未被已有设施点服务半径覆盖的最佳设施点建立位置。实验结果表明,该新型选址模型相较于最小化阻抗模型与最大化覆盖模型,新增优化设施点使整体服务半径覆盖率分别高出10.42%和6.95%,适合求解较为精确且小规模空间下的选址问题。

基于24Model的煤炭企业安全文化提升实践研究

为了发挥安全文化在事故预防中的重要作用,运用基于事故致因“2-4”模型的安全文化水平提升方法研究某煤炭企业的安全文化建设现状并提出改进措施。首先,明确了24Model视域下的安全文化水平提升方法在我国的适用性;然后,从安全文化水平现状和安全理念2方面开展了针对该企业的实践研究;最后,提出改进措施。结果表明:该企业安全文化水平总体上要比国内或同行业的安全业绩较差的企业好,比安全业绩较好的企业差;该企业安全理念涉及32元素的13个方面;一线人员、班组长、专业人员和领导层对部分元素的理解程度不佳,并且这些元素中仅有元素3、元素4、元素6、元素13、元素16的内容在企业的安全理念有所体现。提出了14条改进的安全理念用于提高企业的安全文化水平。

Evaluating Privacy Leakage and Memorization Attacks on Large Language Models (LLMs) in Generative AI Applications

The recent interest in the deployment of Generative AI applications that use large language models (LLMs) has brought to the forefront significant privacy concerns, notably the leakage of Personally Identifiable Information (PII) and other confidential or protected information that may have been memorized during training, specifically during a fine-tuning or customization process. We describe different black-box attacks from potential adversaries and study their impact on the amount and type of information that may be recovered from commonly used and deployed LLMs. Our research investigates the relationship between PII leakage, memorization, and factors such as model size, architecture, and the nature of attacks employed. The study utilizes two broad categories of attacks: PII leakage-focused attacks (auto-completion and extraction attacks) and memorization-focused attacks (various membership inference attacks). The findings from these investigations are quantified using an array of evaluative metrics, providing a detailed understanding of LLM vulnerabilities and the effectiveness of different attacks.

基于Modelica的变工况下汽轮机蒸汽参数建模

汽轮机仿真模型能辅助研究汽轮机蒸汽参数在不同工况下的变化情况,为故障检测与诊断提供先验知识。基于物理机理建立了变工况下汽轮机蒸汽变化的数字仿真模型,对汽轮机变工况条件下的蒸汽参数变化趋势进行了分析。研究结果表明,基于物理机理建立的模型能够较为准确地预测汽轮机在不同工况下的性能表现,仿真模型的结果与实际情况相符,表明该模型具有较好的应用性和可靠性。研究结果对汽轮机的设计和优化具有一定的参考价值,同时也可为汽轮机的运行和维护提供一定的理论支持。

PID控制的阀控缸系统的Modelica建模与仿真实现

针对PID控制的阀控缸系统基于Modelica语言在开源仿真环境OpenModelica中进行了建模和仿真。给出了一套完善的阀控缸系统的建模思路,且所建立的模型具有很好的模型复用性。通过与AMESIM中建立的相同系统在相同参数下的仿真结果进行对比。结果表明,二者仿真结果基本一致,验证了所建立模型的正确性。由于整套系统均在开源环境下建模并仿真,因此解决了在实际工程中仿真PID控制的阀控缸系统时由于商业软件价格昂贵而造成的仿真成本过高的问题。

Constraints on Asymmetric Dark Matter in Quintessence Model

The modified cosmology like quintessence model with kination phase predicted the Hubble expansion rate of the universe before Big Bang Nucleosynthesis is different from the standard cosmological scenario. The modified expansion rate leaves its imprint on the relic density of asymmetric dark matter. In this work, we review the calculation of relic density of asymmetric WIMP dark matter in the standard cosmological scenario and quintessence model with kination phase. Then we use the Planck data to find constraints on the annihilation cross section and the mass of the asymmetric dark matter in those models.

Mitigating the Prevalence of Diabetic Retinopathy in the United States: Utilization of the Chronic Care Model as a Public Health Framework

As the prevalence of diabetic retinopathy continues to be on the rise, the Chronic Care Model (CCM) offers a transformative, patient-focused approach for efficient diabetic retinopathy care, emphasizing the need for urgent and innovative strategies in the United States. The model integrates community resources, healthcare organizations, self-management support, delivery system design, decision support, and clinical information systems. Addressing challenges and solutions, the model emphasizes proactive and preventive measures, collaborative multidisciplinary care, technological integration, and overcoming resistance to change. This paper proposes the utilization of the Chronic Care Model (CCM) as a possible public health framework for comprehensive management of diabetic retinopathy in the United States. Implementing the CCM offers a comprehensive approach to diabetic retinopathy care, addressing both individual and systemic factors, essential for improving public health outcomes.

Simplified Applicable Model for Fire Tube Boiler

This research aims to present a simplified mathematical model to predict the performance of fire tube boilers, taking into account the necessity of knowing the components of exhaust gases and the extent of their compatibility with environmental laws and requirements. The model shown is for a horizontal, three-pass, wet-back fire tube boiler at steady-state, steady-flow operation. It is concluded from the applicability of the model for different boiler capacity ratings that the results are simplified and important for the boiler manufacturers to predict the performance and make the choice to modify the proposed design to achieve certain needs.

基于PSModel的海上风电场全电磁暂态建模与仿真

开发有效的风机电磁暂态模型是进行海上风电并网研究的基础。根据不同生产厂家所提供的直驱风机模型,提出了一种基于厂家黑盒模型的直驱风机电磁暂态建模方法。利用厂家所提供的模型测试数据解析风机的故障响应特性,推导了不同厂家在电压故障下有功和无功功率响应表达式,包括故障期间的穿越控制过程及不同过程间的暂态切换策略。提出了加权平均压降的风电场等值方法,采用自主研发的全电磁暂态仿真软件(power system model,PSModel)对我国广东某海上风电场进行全电磁暂态建模。根据稳态潮流及暂态特性结果验证了模型可通过系统测试,且有效实现了海上风电场的全电磁暂态建模,从而为该风场接入大电网后的安全稳定分析提供了研究基础。

Methods to Reduce the Hypoglycemic Mortality of Alloxan in Diabetic Rabbit Model

Objective: To explore an intervention method to reduce the mortality of alloxan diabetes model, and to preliminarily analyze the mechanism of alloxan induced animal death. Methods: Healthy New Zealand rabbits were randomly divided into injection group, control group, experimental group and blank group. The single injection group was injected with 100 mg/kg alloxan once. The control group was given 5% glucose solution and 100 mg/kg alloxan was injected in two times. The experimental group was given 5% glucose solution orally, 100 mg/kg alloxan, 7 mL 0.9% NaCl intravenously and 5 mL 5% glucose intraperitoneally immediately, and blood glucose was continuously monitored, 10 mL 5% glucose intravenously and 10 mL 5% glucose intraperitoneally every 4 h in the hypoglycemic stage. The blank group does nothing. Liver and kidney tissues at different time periods were stained with HE and organ index was evaluated. Results: 1) A single injection of 100 mg/kg alloxan without any intervention resulted in 100% mortality. Before modeling, oral administration of 5% glucose solution, divided into two injections of 100 mg/kg alloxan, mortality reached 100%;A single injection of 100 mg/kg alloxan and continuous intervention of normal saline and glucose for 20 h can significantly reduce the mortality of alloxan induced diabetic rabbit model. 2) Liver and kidney tissues were damaged in different degrees at different time periods, and liver and kidney indexes were significantly increased after alloxan injection compared with the normal group, with statistical significance (P > 0.05). Conclusion: 1) Every 4 hours of hypoglycemia, 10 ml 5% glucose was injected intravenously 10 ml 5% glucose intraperitoneally. It can reduce the death rate of alloxan diabetic rabbit model and shorten the time of blood glucose measurement. 2) After the injection of alloxan, acute lesions of liver and kidney may occur in different degrees, or one of the causes of acute death of experimental animals.

Multi-Model Approach for Assessing the Influence of Calibration Criteria on the Water Balance in Ouémé Basin

Hydrological models are very useful tools for evaluating water resources, and the hydroclimatic hazards associated with the water cycle. However, their calibration and validation require the use of performance criteria which choice is not straightforward. This paper aims to evaluate the influence of the performance criteria on water balance components and water extremes using two global rainfall-runoff models (HBV and GR4J) over the Ouémé watershed at the Bonou and Savè outlets. Three (3) Efficacy criteria (Nash, coefficient of determination, and KGE) were considered for calibration and validation. The results show that the Nash criterion provides a good assessment of the simulation of the different parts of the hydrograph. KGE is better for simulating peak flows and water balance elements than other efficiency criteria. This study could serve as a basis for the choice of performance criteria in hydrological modelling.

24 Model在道路运输事故成因中的应用

为预防道路运输事故,基于事故致因“2-4”(24 Model)模型,对2015—2021年发生的215起道路运输事故的行为原因进行研究。从个人层面和组织层面分析导致事故发生的直接原因、间接原因、根本原因及其根源原因。结果表明:发生频次最多的不安全动作是驾驶员超速行驶(占10.67%);22.33%的道路运输事故存在管理者违章安排无资质员工上岗作业;管理者和领导者对57.67%的道路运输事故的发生有重要影响;由人为因素产生的不安全物态占比达71.53%;驾驶员、车辆安全生产管理制度欠缺或执行不到位在根本原因中尤为突出(占20.92%)。基于此,提出道路运输事故预防对策并构建出道路运输系统各因素关系图。

Animal Model of Aortic Valve Calcification: Their Methodology Helps Us Understand Aortic Valve Calcification

Aortic valve calcification disease (CAVD) is the most prevalent degenerative valve disease in humans, leading to significant morbidity and mortality. Despite its common occurrence, our understanding of the underlying mechanisms remains incomplete, and available treatment options are limited and risky. A more comprehensive understanding of the biology of CAVD is essential to identify new therapeutic strategies. Animal models have played a crucial role in advancing our knowledge of CAVD and exploring potential treatments. However, these models have inherent limitations as they cannot fully replicate the complex physiological mechanisms of human CAVD. In this review, we examine various CAVD models ranging from pigs to mice, highlighting the unique characteristics of each model to enhance our understanding of CAVD. While these models offer valuable insights, they also have limitations and shortcomings. We propose that the guide wire model shows promise for future CAVD research, and streamlining the methodology could enhance our understanding and expand the research scope in this field.

Hydrological Modelling of the Casamance River in Its Upstream Section (Basin at Kolda Level) to Predict Its Future States as a Function of Different Stresses

Flow records for stations in the Casamance basin are incomplete. Several gaps were noted over the 1980-2021 study period, making this study tedious. The aim of this study is to assess the potential impact of climate change on the flow of the Casamance watershed at Kolda. To this end, hydrological series are simulated and then extended using the GR2M rainfall-runoff model, with a monthly time step. Projected climate data are derived from a multi-model ensemble under scenarios SSP2-4.5 (scenario with additional radiative forcing of 4.5 W/m2 by 2099) and SSP5-8.5 (scenario with additional radiative forcing of 8.5 W/m2 by 2099). An analysis of the homogeneity of the rainfall data series from the Kolda station was carried out using KhronoStat software. The Casamance watershed was then delimited using ArcGIS to determine the morphometric parameters of the basin, which will be decisive for the rest of the work. Next, monthly evapotranspiration was calculated using the formula proposed by Oudin et al. This, together with rainfall and runoff, forms the input data for the model. The GR2M model was then calibrated and cross-validated using various simulations to assess its performance and robustness in the Casamance watershed. The version of the model with the calibrated parameters will make it possible to extend Casamance river flows to 2099. This simulation of future flows with GR2M shows a decrease in the flow of the Casamance at Kolda with the two scenarios SSP2-4.5 and SSP5-8.5 during the rainy period, and almost zero flows during the dry season from the period 2040-2059.