风幕阻止飞机喷漆厂房内可燃气体扩散的数值模拟

应用FLUENT软件中的大涡模拟模型对飞机喷漆厂房内风幕的有效性进行了研究。以某一飞机喷漆厂房内的局部为原型构建网格模型,针对若干种工况,对甲苯蒸气的扩散过程进行了数值模拟,主要研究了甲苯量和风幕速度对甲苯扩散的影响规律。对计算结果进行分析,得到了不同工况下能有效阻止甲苯扩散的临界风幕速度,研究表明在飞机喷漆厂房内设置风幕,只要风幕速度不低于该种工况下的临界风幕速度,就可有效阻止甲苯扩散,降低厂房内可燃气体的爆炸危险性。     

Municipal solid waste generation in Kathmandu, Nepal

Waste stream characteristics must be understood to tackle waste management problems in Kathmandu Metropolitan City (KMC), Nepal. Three-stage stratified cluster sampling was used to evaluate solid waste data collected from 336 households in KMC. This information was combined with data collected regarding waste from restaurants, hotels, schools and streets. The study found that 497.3 g capita(-1) day(-1) of solid waste was generated from households and 48.5, 113.3 and 26.1 kg facility(-1) day(-1) of waste was generated from restaurants, hotels and schools, respectively. Street litter measured 69.3 metric tons day(-1). The average municipal solid waste generation rate was 523.8 metric tons day(-1) or 0.66 kg capita(-1) day(-1) as compared to the 320 metric tons day(-1) reported by the city. The coefficient of correlation between the number of people and the amount of waste produced was 0.94. Key household waste constituents included 71% organic wastes, 12% plastics, 7.5% paper and paper products, 5% dirt and construction debris and 1% hazardous wastes. Although the waste composition varied depending on the source, the composition analysis of waste from restaurants, hotels, schools and streets showed a high percentage of organic wastes. These numbers suggest a greater potential for recovery of organic wastes via composting and there is an opportunity for recycling. Because there is no previous inquiry of this scale in reporting comprehensive municipal solid waste generation in Nepal, this study can be treated as a baseline for other Nepalese municipalities.     

顾及光滑拟合精度的地形曲线采样点布设规则研究

以加权平均抛物线拟合法为例,研究了顾及后期地形曲线光滑模型的采样点布设规则的确定问题.研究表明,对需要光滑拟合处理的地形曲线,除了传统的适当增加冗余采样点(采集质量检测点)外,还需要依据后期的光滑拟合模型,合理规划这些检测点的布点位置,才能有效保障地形曲线的科学采样与高保真重构.     

基于信息扩散的城市区域风险研究

应用信息扩散原理,通过扩散函数对不完备信息进行适当的信息膨胀,建立观测样本和风险概率分布的映射关系,构建了基于信息扩散的城市区域安全生产风险计算模型.并以某市新北区的安全生产风险作为研究对象,选取该区域2005-2010年各年度建筑施工事故、火灾事故、工矿商贸事故的死亡人数和地区亿元GDP死亡率作为该区域安全生产风险的...     

Passive monitoring of nitrogen dioxide in urban air: a case study of Durban metropolis, South Africa

To devise and implement strategies to manage the quality of urban air, a metropolis needs air pollution data on which an air quality management plan can be formulated. Although air pollutants can come from several sources, many reports suggest that nitrogen dioxide from motor vehicle emissions is the major contributor to air pollution in cities. Since vehicles stop or move slowly through traffic intersections, concentrations of nitrogen dioxide (NO(2)) are expected to be relatively high at these sites. Inexpensive Ogawa passive samplers were placed at selected traffic intersections in the Durban Metropolis to trap the NO(2) which was then analysed by a sensitive laboratory-based method. The data obtained by this method was compared with data from sophisticated system comprising an active sampler cum on-line chemiluminescence detector. The sampling was done over a twelve month period to cover all seasons. Statistical analysis of the data showed that there was no significant difference between the means for the two methods. This study has established that an Ogawa passive sampler may be used as an economical and reliable collector for NO(2) in ambient air under varying climatic conditions. Further, the analysis method using a UV-Visible spectrophotometer was sensitive enough to detect NO(2) at the 10-20 ppb level. The cost of the method should be well within the budgets of most municipalities and it would motivate them to develop policies to alleviate traffic congestion.     

基于漂浮箱法和扩散模型法测定淡水养殖鱼塘甲烷排放通量的比较

本研究以我国东南部地区淡水养殖鱼塘为研究对象,于2017年9月到2018年8月采用漂浮箱法和扩散模型法同步原位观测其CH_4排放通量,旨在明确运用两种不同方法观测CH_4的排放特征、排放强度及其驱动因子,综合比较两种方法观测结果的差异性,其中扩散模型法能够进一步量化扩散传输对CH_4排放通量的贡献.结果表明,两种方法观测的CH_4排放通量有相似的季节变化特征,即夏秋季排放高,冬春季排放低.通过漂浮箱法观测淡水养殖鱼塘CH_4排放通量的变化范围为0. 14～3. 13 mg·(m~2·h)~(-1),其年平均排放通量为(0. 86±0. 30) mg·(m~2·h)~(-1),而由扩散模型法估算出鱼塘CH_4排放通量变化范围为0. 04～1. 41 mg·(m~2·h)~(-1),其年平均排放通量为(0. 45±0. 08) mg·(m~2·h)~(-1).基于两种方法观测的CH_4排放通量具有相同的环境驱动因子,CH_4排放通量与水温、底泥可溶性有机碳(DOC)和水体化学需氧量(COD)呈现显著的正相关关系,与水体溶解氧(DO)呈现出极显著的负相关关系.综合比较两种方法观测结果,发现由扩散模型法估算出的淡水养殖鱼塘CH_4排放通量约为漂浮箱法测定结果的45%左右(P 0. 01),扩散模型法可能低估淡水养殖系统CH_4排放通量.综上所述,漂浮箱法更适合用于观测我国东南部内陆地区淡水养殖生态系统CH_4排放.     

两种沉水植物对上覆水和间隙水中可溶性无机氮的影响

在实验室模拟研究沉水植物金鱼藻(Ceratophyllum demersum)、苦草(Vallisneria spiralis L.)对上覆水、间隙水中可溶性无机氮(dissolved inorganic nitrogen,DIN)的浓度、赋存形态及DIN扩散通量的影响.结果表明,金鱼藻和苦草对上覆水中DIN的去除效果均强于间隙水,对各形态DIN的去除差异:NO-2-N>NH+4-N>NO-3-N;金鱼藻对上覆水中DIN的去除效果明显强于苦草,但对间隙水中DIN的去除效果弱于苦草;金鱼藻和苦草减少了NH+4-N和NO-2-N的扩散通量,显著增加了NO-3-N的扩散通量,使NO-3-N取代NH+4-N成为间隙水向上覆水中扩散DIN的主要形态,但对NO-3-N和NH+4-N扩散通量的影响,金鱼藻和苦草之间差异不显著;金鱼藻和苦草增加了上覆水和间隙水中3种形态DIN含量比例的变化幅度,金鱼藻对上覆水中DIN含量比例影响强于苦草,对间隙水中DIN含量比例的影响弱于苦草.总体来讲,金鱼藻对上覆水中DIN形态影响较大,苦草对间隙水中DIN形态影响较大,两者对DIN扩散通量的影响差别不显著.     

水生生物基准推导中物种选择方法研究

运用Mann-Whitney U检验法对不同生物分类群内的物种对氨氮的毒性敏感性进行分析和统计检验,并基于毒理学、生物分类学,从抽样-推断的角度对物种选择进行讨论,并对物种选择与基准推导方法的关系进行梳理.结果发现,物种毒性数据中脊椎动物占绝大部分,尤其是辐鳍纲,与真实生态系统中物种类群结构不符;大部分不同分类群间物种对氨氮的敏感性差异显著,随着物种分类阶元级别降低,大部分分类群内物种的差异逐渐减小,个别分类群内物种的敏感性未呈现相似的规律,尤其是相对不敏感的物种;总体上,能在一定程度上说明同一分类群内的生物有相对接近的敏感性.由抽样-推断理论与方法可知,基准推导属于基于抽样设计的统计推断,因此,可以将生物分类类群作为物种选择的辅助变量,依据水生态系统中物种组成,运用分层随机抽样技术进行物种选择,以提高抽样效率和精度.     

海南岛地区大气输送和扩散特征的数值模拟

采用中尺度气象模式WRF及风场诊断模式CALMET,结合轨迹分析和拉格朗日随机游走模拟方法,分析了海南岛地区低层大气中尺度水平输送和扩散特性,并计算了各季平均大气扩散模态.结果表明,该区域大气污染物的扩散和输送主要受到大尺度背景环流、海陆风等局地环流及地形绕流等的影响.海陆风局地环流是沿海城市源排放的大气污染物向海南本岛输送和扩散的主要机制.北部城市海口的大气扩散对岛内影响最大,冬季平均影响范围可覆盖西北半部;春、秋季主要影响西北和北部区域;夏季对本岛的影响仅限于北部沿岸.南部城市三亚的大气扩散对岛内影响较小,秋季向西南海面的扩散对本岛几乎没有影响;冬、春季对三亚以西沿岸的影响有所增加;夏季扩散影响全面指向岛内,并因地形的作用而东、西向大角度扩展,影响海岛南部的大部分沿岸地区.西北部昌江的平均输送扩散方向与当地海岸线的走向基本一致,污染影响不易深入到岛内.其中,秋季扩散影响以偏西南方向为主,仅对昌江西南部分海岸有少量影响;冬、春季扩散形态类似但影响范围扩大到以东方市为代表的低山盆地地带;海岛西北部大部分沿海地带可受到昌江夏季扩散的影响,但平均扩散方向指向东北偏北的海面.     

基于人工神经网络的烟气及温度实时预测模型*

为监测建筑火灾事故区域的危险程度,实现更加安全、高效的火灾应急救援,以通廊式建筑为研究对象,基于转置卷积神经网络及数值模拟方法开发1种可实时预测走廊位置处烟气扩散和温度分布的神经网络模型。首先,依托Python建立包含全连接、转置卷积、反池化等在内的19层神经网络模型的整体架构;其次,建立包含99个火灾场景,共7 920组图像数据的火场信息数据库用于模型训练;最后,使用测试集对模型进行可靠性验证。研究结果表明:烟气（温度）预测模型在不同火灾场景下的预测精度达到95%,训练完成后模型的预测时间一般为1~2 s。研究结果可为应急策略的快速制定提供数据参考。