环境决策支持系统中两类模型方法的整合

该文对环境决策支持系统中情景分析和数学优化两类模型方法的优缺点和适用范围进行了比较分析。实例研究表明,两类模型方法的整合使用可以有效地支持较为复杂的决策过程。      

生物载体强化的连续流生物制氢反应器的运行特性

在连续流搅拌槽式反应器(CSTR)中填加比重为1.54 g/cm³,粒径小于2mm的多孔物质,以糖蜜废水为底物利用活性污泥制取氢气.考察了填加生物载体后生物制氢反应器连续流稳定运行的系统特性.研究表明,投加生物载体能够扩大产氢细菌的活性范围,提高系统的抗冲击负荷能力和耐低pH值的能力,增加系统稳定性,并且可使系统在低HRT下保持较高的生物量.此连续流生物制氢反应系统的最佳发酵类型为乙醇型发酵,适宜的pH值范围为3.8～4.4,气相中的氢气含量约为40%～57%,最大产氢速率为0.37L/(g·d).降低pH值可抑制厌氧发酵过程中出现的产甲烷菌群,加速产氢反应器的启动.     

我国普及推广循环流化床锅炉的必要性及推广设想

循环流化床锅炉在环保、节能等方面效果显著,在我国推广应用有重大意义,文章对该技术在我国推广普及的必要性及基本原则和思路进行了论述.     

Development of a Decision Support System for Assessing Farm Animal Welfare in Relation to Husbandry Systems: Strategy and Prototype

Due to increasing empiricalinformation on farm animal welfare since the1960s, the prospects for sound decisionmakingconcerning welfare have improved. This paperdescribes a strategy to develop adecision-making aid, a decision support system,for assessment of farm-animal welfare based onavailable scientific knowledge. Such a decisionsupport system allows many factors to be takeninto account. It is to be developed accordingto the Evolutionary Prototyping Method, inwhich an initial prototype is improved inreiterative updating cycles. This initialprototype has been constructed. It useshierarchical representations to analysescientific statements and statements describingthe housing system. Welfare is assessed fromwhat is known about the biological needs of theanimals, using a welfare model in the form of atree that contains these needs as welfarecomponents. Each state of need is assessedusing welfare relevant attributes of thehousing system and weighting factors.Attributes are measurable properties of thehousing system. Weighting factors are assignedaccording to heuristic rules based on theprinciple of weighting all components(attributes and needs) equally, unless thereare strong reasons to do otherwise. Preliminarytests of the prototype indicate that it may bepossible to perform assessment of farm-animalwelfare in an explicit way and based onempirical findings. The procedure needs to berefined, but its prospects are promising.     

Process conditions of preparing methanol from cornstalk gas

The low-heat-value cornstalk gas produced in the down-flow fixed bed gasifier was tentatively used for methanol synthesis. The cornstalk gas was purified and the technical procedures such as deoxygenation, desulfurization, catalytic cracking of tar, purification and hydrogenation were studied. The catalytic experiments of methanol synthesis with cornstalk syngas were carried out in a tubular-flow integral and isothermal reactor. The effect of reaction temperature, pressure, catalysttypes, catalyst particle size, syngas flow at entering end and composition of syngas was investigated. The optimum process conditions and yield of methanol from cornstalk syngas were obtained. The experimental results indicated that the proper catalyst of the synthetic reaction was C301 and the optimum catalyst size (φ) was 0.833 mm×0.351 mm. The optimum operating temperature and pressure were found to be 235℃ and 5 Mpa, respectively. The suitable syngas flow 0.9-1.10 mol/h at entering end was selected and the best composition of syngas were CO 10.49%, CO2 8.8%, N2 37.32%, CnHm 0.95% and H2 40.49%. The best methanol yield is 0.418 g/g cornstalk. The study provided the technical support for the industrial test of methanol production from biomass (cornstalk)gas.     

基于规范变换与误差修正的回归支持向量机的环境系统预测

为了建立不同环境系统皆能规范、统一、简洁、实用的回归支持向量机预测模型,针对传统的回归支持向量机预测模型存在结构不能普适、规范和统一及用于大样本、多因子预测会出现学习效率低、求解精度差的局限,提出适用于环境系统预测量及其影响因子参照值和规范变换式的设计原则和方法,使规范变换后的影响因子皆"等效"于同一个规范影响因子;为提高样本的预测精度,还提出预测样本的模型输出的误差修正法.在对环境系统的预测量及其影响因子进行规范变换的基础上,由有m个规范影响因子的每个建模样本生成m个"等效"训练样本,从建模样本中,选择各影响因子的最大规范值组成训练样本集的"参考样本",计算核函数中每个训练样本相对于"参考样本"的范数;并应用优化算法优化模型参数,建立适用于预测量及其影响因子规范值的仅有2个或3个支持向量的两种简单结构的回归支持向量机预测模型.将基于规范变换的两种简单结构的回归支持向量机模型与相似样本误差修正法相结合,用于河津大桥监测断面6个样本的COD月平均值预测,并与多种传统预测模型和方法进行了比较.结果表明:对同一个预测样本,两种模型的预测值十分接近;此外,两种预测模型用于6个样本预测,其相对误差绝对值的平均值分别为2.09%、2.79%,均远小于传统的投影寻踪回归预测的41.63%、支持向量机预测的40.99%、灰色神经网络预测的25.94%和马尔可夫预测的10.16%;而两种预测模型对异常样本预测的最大的相对误差绝对值分别为5.85%、5.13%,更加远远小于传统的4种预测模型的169.07%、180.45%、68.44%、41.96%.两种基于规范变换的回归支持向量机预测模型简洁、普适、规范和统一,避免了"大样本数困难",提高了学习效率和模型的预测精确度,对其他预测建模法也有借鉴作用.     

基于小波分解与自适应多级残差修正的最小二乘支持向量回归预测模型的PM2.5浓度预测

采用小波分解(WD)将济南市科干所监测站PM_(2.5)浓度的一维时间序列(2013年1月1日—2017年8月15日)分解为高维信息,获得了该监测站附近PM_(2.5)浓度的时频变化特征,重点分析了PM_(2.5)的随机性和趋势性问题.然后构建了基于小波分解的多级残差修正的最小二乘支持向量回归预测模型(AMLRC-WLSSVR),结果发现,该模型能够很好地对济南市PM_(2.5)浓度做出预测,特别是针对重污染天气的预测有很好的精度.为了避免预测结果的不确定性问题,提出了一种基于方差估计给出预测值置信区间上界的方法,同时,有效弥补了单点预测的不稳定性及预测精度不足的缺点,该方法能够为实际空气污染预警提供技术支持.     

A microcomputer-based emergency response system*

A microcomputer-based system was developed to provide local officials responsible for disaster management with assistance during the crucial period immediately following a disaster, a period when incorrect decisions could have an adverse impact on the surrounding community.
While the paper focuses on a potential disaster resulting from an accident at a commercial nuclear power generating facility, the system can be applied to other disastrous situations. Decisions involving evacuation, shelter and the deployment of resources must be made in response to floods, earthquakes, accidents in the transportation of hazardous materials, and hurricanes to name a few examples.
As a decision aid, the system was designed to enhance data display by presenting the data in the form of representations (i.e. road maps, evacuation routes, etc.) as well as in list or tabular form. The potential impact of the event (i.e. the release of radioactive material) was displayed in the form of a cloud, representing the dispersion of the radioactive material. In addition, an algorithm was developed to assist the manager in assigning response resources to demands.
The capability for modelling the impact of a disaster is discussed briefly, with reference to a system installed in the communities surrounding the Indian Point nuclear power plant in New York State.
Results demonstrate both the technical feasibility of incorporating microcomputers indecision support systems for radiological emergency response, and the acceptance of such systems by those public officials responsible for implementing the response plans.     

Evacuation behavior and Three Mile Island

The responses of the residents to the nuclear power plant arcident at Three Mile Island, Pennsylvania illustrate the factors influencing pre-impact coping responses of populations exposed to technological hazards. Confusion itnd ambiguous information influenced both the decision to evaluate and to remain in place. Proximity to the facility, stage in life cycle and the actions of friends and neighbors influenced the decision to evacuate.     

蜂窝状活性炭研究

蜂窝状活性炭是一种新型的活性炭。简要介绍了蜂窝状活性炭的结构特性和吸附性能，重点总结了蜂窝状活性炭基催化剂的制备方法及其在催化方面的应用。