Risk assessment and risk reduction of an acrylonitrile production plant

Reducing accident occurrence in petrochemical plants is crucial, thus appropriately allocating management resources to safety investment is a vital issue for corporate management as international competition intensifies. Understanding the priority of safety investment in a rational way helps achieve this objective.In this study, we targeted an acrylonitrile plant. First, Dow Chemical's Fire and Explosion Index (F&EI) identified the reaction process as having the greatest physical risk. We evaluated the severity of accidents in the reaction process using the Process Safety Metrics advocated by the Center for Chemical Process Safety (CCPS); however, this index does not express damages a company actually experience. To solve this problem, we proposed a new metric that adds indirect cost to CCPS metrics. We adopted fault tree analysis (FTA) as a risk assessment method. In identifying top events and basic events, we attempted to improve the completeness of risk identification by considering accidents from the past, actual plant operation and equipment characteristics, natural disasters, and cyber-attacks and terrorist attacks. Consequently, we identified the top events with high priority in handling because of serious accidents as fire/explosion outside the reactor, fire/explosion inside the reactor, and reactor destruction. The new CCPS evaluation index proposed in this study found that fire and explosion outside the reactor has the highest severity. We considered the creation of the fault tree (FT) diagram of the top event, estimating the occurrence probability, and identifying the risk reduction part and capital investment aimed at risk reduction. As an economically feasible selection method for risk reduction investment, using the difference in loss amounts before and after safety investments indicated investment priority.     

面向实时人群动力学分析的深度基本图网络

当前行人疏散实验中基本图计算方法通常是通过对每个行人进行跟踪实现的.但这种跟踪方法难以实现实时人群动力学分析.针对这一问题,提出了深度基本图网络.实验提出的网络框架由两个模块组成,即多尺度递归卷积神经网络(MSR-Net)和光流模块,分别对行人密度和行人速度进行估计.具体来讲,MSR-Net学习了输入图像与行人密度图之...     

基于CA的城市形态扩展多解模拟——以北京市东部平原区情景分析为例

构建了一个基于人工神经网络的约束型城市扩展CA模型。利用该模型,预测了北京市东部平原区在3种情景规划之下的未来50年的城市形态,为不同的城市发展模式之间的比较分析奠定基础。然后以2024年的北京东部平原区模拟城市形态为基准,从自然生态功效、社会服务功效、经济利益功效3个方面对3种规划模式进行了情景分析,从而系统的比较了3种规划模式的复合生态功效,为城市规划决策提供有力的支持。     

半导体企业的EHSMS评价体系研究

为了给员工提供安全、环保、健康的工作环境,半导体制造企业必须建立环境管理和职业健康安全管理体系(EHSMS).为此,评价EHSMS运行的有效性、完整性以及合法性就显得十分重要.本文在分析半导体企业污染物影响环境和健康基础上,建立了EHSMS评估体系.为了实施EHSM评估系统,方便半导体企业用户使用,提出了一套EHSM评价系统的计算机软件解决方案.     

小波网络在大气污染物浓度预测中的应用

提出一种“分解-重构-预测”小波网络的大气污染物浓度的预测方法。通过小波分解,把浓度序列分解为不同频段的小波系数序列,再对各层的小波系数子序列重构到原尺度上,然后对小波系数序列采用相匹配的BP神经网络模型进行预测,最后合成浓度序列的最终预测结果。经对二氧化硫浓度预测证明,该方法预测模型推广能力强,预测精度高。     

BP神经网络在台风路径预报中的应用

利用前馈型BP神经网络模型,对发生于中国沿海的热带气旋的移动路径进行了预报应用研究.根据中国<台风年鉴>发布的每个台风过程记录,对预报试验的台风个例分别选取了经度、纬度、中心气压和最大风速等81个因子,由多元回归选取了其中相关性好的因子,进行网络的学习训练,在获取前24h间隔6h的4次台风信息的基础上,用来预报了台风未来24h,48h和72h的短期路径变化.将该方法预报结果与CLIPER模式预报结果进行了比较,结果表明,BP神经网络模式的预报精度比CLIPER模式的高.     

Anaerobic tapered fluidized bed reactor for starch wastewater treatment and modeling using multilayer perceptron neural network

Anaerobic treatability of synthetic sago wastewater was investigated in a laboratory anaerobic tapered fluidized bed reactor(ATFBR) with a mesoporous granular activated carbon(GAC)as a support material.The experimental protocol was defined to examine the effect of the maximum organic loading rate(OLR),hydraulic retention time(HRT),the efficiency of the reactor and to report on its steady- state performance.The reactor was subjected to a steady-state operation over a range of OLR up to 85.44 kg COD/(m~3.d).The COD removal efficiency was found to be 92% in the reactor while the biogas produced in the digester reached 25.38 m~3/(m~3·d) of the reactor. With the increase of OLR from 83.7 kg COD/(m~3.d),the COD removal efficiency decreased.Also an artificial neural network(ANN) model using multilayer perceptron(MLP)has been developed for a system of two input variable and five output dependent variables. For the training of the input-output data,the experimental values obtained have been used.The output parameters predicted have been found to be much closer to the corresponding experimental ones and the model was validated for 30% of the untrained data.The mean square error(MSE)was found to be only 0.0146.     

河流综合水质评价方法比较研究

为探讨综合水质评价结论的科学合理性,对典型的7种河流综合水质评价方法,包括单因子评价法、污染指数法、模糊评价法、灰色系统评价法、层次分析法、人工神经网络法、水质标识指数法,在归类分析其特性的基础上,针对代表性的评价样本,进行了综合水质评价比较研究。主要结论为：①单因子评价法以最差水质指标所属类别作为综合水质类别,评价结论表现为过保护;②污染指数法能够直观判断综合水质是否达到功能区目标,但是不能判断综合水质类别;③对模糊数学法、灰色系统评价法、层次分析法、BP人工神经网络法和水质标识指数法等5种典型评价方法,当综合水质为Ⅰ~Ⅴ类水情形时,这几种方法评价结论基本一致,证明这5种典型评价方法对Ⅰ~Ⅴ类水的评价结论具有科学合理性;④当综合水质为劣Ⅴ类水时,模糊数学法、灰色系统评价法、层次分析法、BP人工神经网络法的评价结论偏保守,而水质标识指数法解决了劣Ⅴ类水质的连续性描述问题,能够对劣Ⅴ类水进行科学合理评价,并对Ⅰ~Ⅴ类水、劣Ⅴ类水不黑臭、劣Ⅴ类水黑臭进行全面、科学合理的定性与定量评价.     

基于BP神经网络的鄱阳湖水位模拟

考虑到鄱阳湖水位受流域五河与长江来水等多因素的共同作用而表现出高度非线性响应,采用典型的三层BPNN神经网络模型来模拟鄱阳湖水位与其主控因子之间的响应关系。分别将湖口、星子、都昌、棠荫和康山水位作为目标变量进行BPNN模型构建和适用性评估。结果显示：综合考虑流域五河及长江来水（汉口或九江）的BPNN水位模型,空间站点水位模拟精度（R2和Ens）可达090以上,各站点的均方根误差（RMSE）变化范围约050～10 m,若忽略长江来水的影响作用,仅将流域五河来水作为湖泊水位的主控影响因子,模型训练期与测试期的纳希效率系数（Ens）和确定性系数（R2）显著降低,且低于050,均方根误差（RMSE）也明显增大（124～288 m）,意味着综合考虑流域五河与长江来水是获取结构合理、精度保证的鄱阳湖水位模型的重要前提。同时建议针对鄱阳湖湖盆变化对水位的影响,尽可能选择一致性较好的长序列数据集来训练和测试BPNN模型。所构建的BPNN神经网络模型可进一步结合流域水文模型,用来预测气候变化与人类活动下流域径流变化对湖泊水位的潜在影响,也可作为一种有效的模型工具来回答当前鄱阳湖一些备受关注的热点问题,如定量区分流域五河与长江来水对湖泊洪枯水位的贡献分量,为湖泊洪涝灾害的防治和对策制定提供科学依据     

Neural network forecasting of air pollutants hourly concentrations using optimised temporal averages of meteorological variables and pollutant concentrations

The new method for the forecasting hourly concentrations of air pollutants is presented in the paper. The method was developed for a site in urban residential area in city of Zagreb, Croatia, for four air pollutants (NO₂, O₃, CO and PM₁₀). Meteorological variables and concentrations of the respective pollutant were taken as predictors. A novel approach, based on families of univariate regression models, was employed in selecting the averaging intervals for input variables. For each variable and each averaging period between 1 and 97 h, a separate model was built. By inspecting values of the coefficient of correlation between measured and modelled concentrations, optimal averaging periods for each variable were selected. A new dataset for building the forecasting model was then calculated as temporal moving averages (running means) of former variables. A multi-layer perceptron type of neural networks is used as the forecasting model. Index of agreement, calculated for the entire dataset including the data for model building, ranged from 0.91 to 0.97 for the respective pollutants. As suggested by the analysis of the relative importance of the input variables, different agreements for different pollutants are likely due to different sources and production mechanisms of investigated pollutants. A comparison of the new method with more traditional method, which takes hourly averages of the forecast hour as input variables, showed similar or better performance. The model was developed for the purpose of public-health-oriented air quality forecasting, aiming to use a numerical weather forecast model for the prediction of the part of input data yet unknown at the forecasting time. It is to expect that longer term averages used as inputs in the proposed method will contribute to smaller input errors and the greater accuracy of the model.