基于遗传算法的支持向量机预测有机物自燃点的研究

根据定量构效关系(QSPR)原理,研究自燃点(AIT)与其分子结构间的内在定量关系。以265种有机化合物作为样本集,随机选择238种作为训练集,27种作为测试集,用遗传算法(GA)进行变量选择,分别建立多元线性回归(MLR)模型和支持向量机(SVM)模型研究有机物的自燃点与其分子结构间的关系。通过分析,发现造成模型预测效果不佳的原因是试验数据本身存在问题。通过对2个模型的比较,结果为GA-SVM模型明显优于GA-MLR模型,说明自燃点与其分子结构间具有很强的非线性关系。     

炼油厂VOCs排放及其对O3生成的敏感性研究

对某炼油石化厂进行了O3体积分数和VOCs的监测分析,获得炼油厂周边地区各月O3的体积分数变化特征和日O3体积分数变化特征.研究发现,该炼油厂地区O3体积分数经常超过国家二级标准,说明该地区O3污染严重.该地区O3日变化曲线不同于所在城市中心的O3变化特征.炼油厂地区O3体积分数在6-8月显示出双峰特性以及更长的最大值持续时间.与此同时,在5月、9月、10月使用SUMMA采样罐在炼油厂厂界进行VOCs样品采集,使用气相色谱质谱联用仪(GC - MS)对样品进行分析,得到厂界VOCs各组分的体积分数,烷烃比例最高,约82.6％,其中直链烷烃占49.9％,而芳香烃和烯烃仅为10.2％和7.5％.VOCs监测结果与文献相近,烯烃排放量略低.此外,为了对炼油石化厂排放的VOCs各物种敏感性特征进行量化描述,利用丙烯等效浓度法,对炼油厂区域各VOCs物种的反应活性进行了归一化研究,量化了该地区各物种的反应活性,获得了综合反应活性较强的5种烯烃、4种直链烷烃和1种环烷烃.     

爆炸形成过程中火焰加速的试验研究

为预防和控制工业爆炸事故,并为脉冲爆轰发动机的研究提供理论指导,分析火焰加速导致的燃烧转爆轰过程的影响因素。采用爆轰管探讨障碍物的阻塞比、混合物的组成、初始压力和点火能等4个因素对爆炸性气体火焰速度和爆轰压力的影响规律。试验结果表明:障碍物的存在能大大提高火焰速度和爆轰压力;爆轰压力随管内障碍物阻塞比的增大先变大后减小,并在阻塞比为0.498,燃料种类为天然气,化学当量比为1时达到最大;爆轰压力还随混合气体初始压力的增大和点火能的提高而增大。选择适宜的条件可大大提高火焰加速速率,促进燃烧向爆轰过程转变。     

民机非包容性转子失效危害防护设计研究

民用飞机的发动机转子非包容性失效给飞机的安全运行带来巨大的风险,为了将这种转子非包容性失效可能带来的危害降至最小,需要合理的评估碎片的危害水平,以及采取合理的安全预防措施。尽管在FAA(Federal Aviation Administration)用于评估碎片危害水平的UEDDAM(Uncontained Engine Debris Damage Analysis Model)软件工具中采用了加强防护设计以减少碎片危害,但是没有对具体的分析流程和碎片能量衰减的计算模型进行说明。为此,在现有的碎片危害分析方法的基础上,考虑到碎片在侵彻过程中会有能量衰减,分析了降低碎片危害的三种安全预防措施,通过对比确定了加强防护设计对于降低碎片危害的重要意义及其分析的流程。通过对飞机结构中关键区域定义的理解,明确了在飞机的不同部位需要采取不同的加强防护设计,并详细介绍了整体加强防护设计和局部加强防护设计的流程。     

武汉市政污泥处理现状及建议

对武汉市现有市政污泥的产生情况和处理作了介绍,并对比了现有的污泥处理技术.结合武汉市的市政污泥处理处置的规划和武汉市的实际情况,提出将来适合武汉市市政污泥处理的技术路线.     

土壤中酞酸酯类化合物监测方法的质量控制研究

从监测结果控制的角度对近10年的土壤中PAEs监测方法进行了探讨,重点比较了索氏萃取柱净化GC/MS、微波萃取柱净化GC、加速溶剂萃取柱净化GC/MS和加速溶剂萃取GPCGC/MS方法的检出限、精密度和准确度。旨在为中国土壤监测方法的建立提供基础支撑,为开展质量控制工作提供评价依据。     

Heavy metal concentration and speciation of seven representative municipal sludges from wastewater treatment plants in Northeast China

The analysis of heavy metals is very important for assessing the feasibility of the agricultural utilization for the municipal sludge. In this paper, a four-step sequential extraction method was applied to extract heavy metals (Cu, Zn, Mn, Cr, and Ni) in municipal sludges from seven individual wastewater treatment plants located in Jilin and Heilongjiang Province, China, for estimating the mobility and bioavailability of the metal ions in the agricultural application. The total concentrations of heavy metals and their chemical fractions after the sequential extraction were determined. Principal component analysis (PCA) was applied to analyze the relations of heavy metals fractions in the municipal sludges. Experimental results indicated that the total concentrations of Cu, Zn, Cr, and Ni in all sludge samples were below the threshold values set out by the Chinese legislation (GB18918-2002). Specially, Zn had a high bioavailability and mobility, Cu and Cr had potential bioavailability, while Mn mainly existed in the residual fraction of municipal sludge. On the other hand, Ni had different mobility in different municipal sludge. PCA results were confirmed by the environmental behavior of heavy metals.     

Spatial and seasonal distribution of nitrogen in marsh soils of a typical floodplain wetland in Northeast China

Horizontal and profile distributions of nitrogen in marsh soils in different seasons were studied in a typical site within the Erbaifangzi wetland in Northeast China. Results showed that there was higher spatial heterogeneity for nitrate nitrogen (NO₃^--_{3}^{-}-N) and ammonium nitrogen (NH₄⁺_{4}^{+}–N), as well as available nitrogen (AN), in surface soils in July compared to that in September. Relative to July, the mean nitrogen contents in surface soils were slightly higher in September; however, in November, soils contained significantly lower NO₃^--_{3}^{-}-N and NH₄⁺_{4}^{+}–N, higher AN, organic nitrogen (Org-N), and total nitrogen (TN). Except for mineral nitrogen, no significant differences were observed between Org-N and TN contents in September and November. Nitrogen contents generally declined exponentially with depth along soil profiles in three sampling dates (July, September, and November), except for a significant accumulation peak of NO₃^--_{3}^{-}-N at the 20–30 cm depth in September. However, NH₄⁺_{4}^{+}–N contents showed a vertical alternation of “increasing and decreasing” in both July and September, while nearly kept constant with depth in November. The depth ranking of nitrogen showed the shallowest distribution for AN, followed by Org-N and TN, while deeper distributions for NO₃^--_{3}^{-}-N and NH₄⁺_{4}^{+}–N. TN, Org-N, and AN were significantly correlated with soil organic matter and total phosphorus. Soil pH values were significantly correlated with TN and AN contents in surface soils. Clay contents showed significant correlations with nitrogen contents except for NO₃^--_{3}^{-}-N in surface soils and NH₄⁺_{4}^{+}–N in profile soils. However, soil moisture was not significantly correlated with nitrogen contents among all soil samples.     

Migration of natural estrogens around a concentrated dairy-feeding operation

Concentrated animal feeding operations have been recognized as one of the most important contributors of natural estrogens which show significant endocrine-disrupting properties in aquatic environments. In this study, the concentrations of 17α-estradiol (17α-E2), 17β-estradiol (17β-E2), estrone (E1), and estriol (E3) in several matrices, including soils (surface and deep), sediments (surface and deep), and groundwaters, around a typical dairy farm were surveyed using gas chromatography/mass spectrometry. Of the two farmlands, surface and subsurface sediments in waste lagoon and along effluent drainage drench, the concentrations of 17α-E2, 17β-E2, and E1 ranged from below detection limit to the highest level of 6.60 μg/kg, except that E3 was not detectable. Three estrogens of 17α-E2, 17β-E2, and E1 with the concentrations of 3.18-31.61 ng/L were observed in two groundwater samples. The results clearly demonstrated the vertical migration and horizontal transport of estrogens in the investigated area. Within 750-m distance, it was observed the attenuation of 17α-E2, 17β-E2, and E1 along the effluent route and the horizontal migration of estrogens was less than 1,350 m in this survey.     

Molecular detection of three gastroenteritis viruses in urban surface waters in Beijing and correlation with levels of fecal indicator bacteria

To assess the presence of three gastroenteritis viruses responsible for human acute gastroenteritis in surface water, a 1-year study was carried out in the city of Beijing, China. A total of 108 urban surface water samples were collected from nine collection sites which were defined with a global positioning system in rivers or lakes from September 2006 to August 2007. The water samples were subjected to virus concentration using an HA electronegative filter, followed by polymerase chain reaction (PCR) for rotavirus (RV) astrovirus (AV), and norovirus (NV). It showed that the number of viruses detected in water samples from different sites was variable, totaling 63 virus strains, with rotavirus (48.1%) verified as the most prevalent detected, followed by astrovirus (AV, 5.6%), and norovirus (NV, 4.6%). RV was also quantified by real-time PCR and the concentration of RV ranged from 0 to 18.27 genome copies·L(-1). And the distributions of RV in surface water were abundant in cold weather (from September to February) while less prevailing in warm weather (from March to August). The high detection rate of RV we encountered in this study provided convincing evidence that RV circulated at a certain frequency in the Beijing population. There was no statistically significant correlation between RV levels and both fecal coliform (R (2)?=?0.02) and Enterococcus faecalis (R (2)?=?0.02) densities. Our study suggests prolonged virus persistence in aquatic environments and emphasizes the enteric virus group as the most reliable for environmental monitoring.