基于社会科学统计程序(SPSS)回归性分析的尾矿库事故预测模型

基于社会科学统计程序(SPSS)软件的分析功能,对调查研究中获取的尾矿库案例进行数据提炼和分类编码,找出相关因子并进行回归性分析。最终目的是找出尾矿库各个因素的内在联系,建立简单的尾矿库事故模型,从而可以初步预测尾矿库事故发生的可能性。该预测模型为尾矿库事故的研究提出了新方法,对于防灾减灾以及保护人民生命财产安全起到了积极作用。     

The accounting method and application of CO2 emissions responsibility by the electricity sector at the provincial level in China

When accounting the CO₂ emissions responsibility of the electricity sector at the provincial level in China,it is of great significance to consider the scope of both producers’ and the consumers’ responsibility,since this will promote fairness in defining emission responsibility and enhance cooperation in emission reduction among provinces.This paper proposes a new method for calculating carbon emissions from the power sector at the provincial level based on the shared responsibility principle and taking into account interregional power exchange.This method can not only be used to account the emission responsibility shared by both the electricity production side and the consumption side,but it is also applicable for calculating the corresponding emission responsibility undertaken by those provinces with net electricity outflow and inflow.This method has been used to account for the carbon emissions responsibilities of the power sector at the provincial level in China since 2011.The empirical results indicate that compared with the production-based accounting method,the carbon emissions of major power-generation provinces in China calculated by the shared responsibility accounting method are reduced by at least 10%,but those of other power-consumption provinces are increased by 20% or more.Secondly,based on the principle of shared responsibility accounting,Inner Mongolia has the highest carbon emissions from the power sector while Hainan has the lowest.Thirdly,four provinces,including Inner Mongolia,Shanxi,Hubei and Anhui,have the highest carbon emissions from net electricity outflow- 14 million t in 2011,accounting for 74.42% of total carbon emissions from net electricity outflow in China.Six provinces,including Hebei,Beijing,Guangdong,Liaoning,Shandong,and Jiangsu,have the highest carbon emissions from net electricity inflow- 11 million t in 2011,accounting for 71.44% of total carbon emissions from net electricity inflow in China.Lastly,this paper has estimated the emission factors of electricity consumption at the provincial level,which can avoid repeated calculations when accounting the emission responsibility of power consumption terminals（e.g.construction,automobile manufacturing and other industries）.In addition,these emission factors can also be used to account the emission responsibilities of provincial power grids.     

广州市白云山降水的化学特征及源解析

以2006年广州市白云山降水监测资料为基础,对其降水样品的化学特征进行了分析。结果表明,白云山降水样品pH值分布为3.04—6.41,雨量加权pH平均值为4.26,酸雨频率为70.2％,降水中的主要阳离子是Ca^2＋和NH^＋4分别占阳离子总量的46.4％和17.5％;主要阴离子为SO^2-4和NO^-3,分别占阴离子总量的59.0％和19.9％。相对酸度（FA）和中和因子（NF）计算结果表明,约有84.7％的降水酸度被碱性成分Ca^2＋和NH^＋4中和。利用富集系数方法,计算结果表明SO^2-4和NO^-3主要来源于人类活动,而Ca^2＋和K^＋主要来源于岩石／土壤风化。     

PSO和SVM混合算法确定太湖入湖河流水质主要影响因子

以影响太湖入湖河流水质的24个因子值为研究对象,将PSO算法与SVM算法相结合。PSO算法用于优化SVM算法的参数c和g,以利于快速、高效地确定c和g的全局最优值;SVM算法基于最优的c和g,分别以24,21,18,15,12,9和6个因子作为特征向量预测水质的污染程度。结果表明,当特征向量为9个影响因子时预测率最高。其参数c=18.56,g=1.35,对应的预测率为：全局预测率92.59%,重度污染水质预测率88.89%,轻度污染水质预测率94.45%。因此,通过PSO和SVM混合算法,可以确定影响太湖入湖河流水质的主要因子,利用这些主要因子对水质进行预测预警,不但可以节省时间,而且可以得到精确的结果。     

Application of Multiattribute Decision-Making Methods for the Determination of Relative Significance Factor of Impact Categories

A relative significance factor (f _i ) of an impact category is the external weight of the impact category. The objective of this study is to propose a systematic and easy-to-use method for the determination of f _i . Multiattribute decision-making (MADM) methods including the analytical hierarchy process (AHP), the rank-order centroid method, and the fuzzy method were evaluated for this purpose. The results and practical aspects of using the three methods are compared. Each method shows the same trend, with minor differences in the value of f _i . Thus, all three methods can be applied to the determination of f _i . The rank order centroid method reduces the number of pairwise comparisons by placing the alternatives in order, although it has inherent weakness over the fuzzy method in expressing the degree of vagueness associated with assigning weights to criteria and alternatives. The rank order centroid method is considered a practical method for the determination of f _i because it is easier and simpler to use compared to the AHP and the fuzzy method.     

外绝缘用液体硅橡胶的自然老化因素研究

近年来,液体硅橡胶(Liquid Silicone Rubber,LSR)已经在外绝缘领域得到了较为广泛的应用,但是目前对其耐老化性能的研究还不够深入。尤其是在中国南方地区使用时,高温、高湿环境更容易导致液体硅橡胶的老化。在许多地区都观察到了电流、电压互感器的液体硅橡胶绝缘外套的龟裂、老化现象,对设备的安全运行产生了影响。本文设计并实施了在光照、高温、高湿、电晕等老化因素作用下的液体硅橡胶人工老化试验,并对老化试验后的样品进行了硬度、力学性能、憎水性、光泽度以及SEM、FTIR等微观性能试验和分析。认为在光照作用下,高温、高湿环境对液体硅橡胶具有加速老化作用,表现为硬度上升、憎水性下降、光泽度下降,并且在表面出现细微裂纹。     

城市化水平测度及其与水环境质量关联性分析

中国城市化正处在快速发展时期,城市化进程对水环境质量产生了重要影响.在对合肥城市化水平进行测度的基础上,利用灰色关联分析法对合肥城市化与巢湖水环境质量进行了关联分析.结果显示：合肥市产业结构对巢湖西半湖水环境质量的影响较大;在城市化发展初期,第二产业成为影响水质的主要因素;近年随着产业结构调整,第三产业比重逐步提高,成为水质改善的驱动力.因此,在合肥城市化进程中第二、三产业比重对水环境的影响不容忽视,应加强产业结构调整,促进城市化与水环境质量的协调发展.     

我国农业面源污染的控制与管理研究

主要通过我国污染源普查和2013年环境公报介绍了我国农业面源污染的现状,并从土壤侵蚀、农业化肥及农药、畜禽粪尿污染、农村生活污水和固体废弃物垃圾、大气干湿沉降、航运污染等方面介绍了我国污染源农业面源污染的成因,结合对土地利用方式、农业活动和田间管理、畜禽粪便及垃圾污染、社会经济因素等农业面源污染的影响因素的分析,分别从技术措施、法律措施和经济措施3方面入手提出了我国面源污染的控制和管理措施。     

生态小区特征和指标体系探讨

保护和建设生态小区，推行生态细胞工程是自然生态工作的有力补充。对于目前出现的许多生态小区，中国尚无一套正式通用的生态小区考评指标体系。本文从生态学的角度出发，探讨了生态小区的概念、内涵与其特征，并以此为基础，提出了生态小区的考评指标体系，希望以此引导四川省生态小区的创建活动。     

青岛市挥发性有机物排放清单及重点行业排放特征研究

根据收集的青岛市九大类排放源的活动水平数据,本研究采用排放因子法结合调研实测等工作建立了青岛市VOCs源排放清单,结果表明,工业企业VOCs排放占总排放的比例达到43.17%。其中,工艺过程源类中排放占比较高的行业依次为橡胶和塑料制品业、非金属矿物制品业、原油加工及石油制品制造业、化学原料和化学制品制造业、黑色金属冶炼和压延加工业等;溶剂使用源类中排放占比较高的行业为金属制品业、皮革皮毛羽毛制品和制鞋业、印刷业、铁路船舶航空等设备制造业、汽车制造业等。通过对重点行业重点企业进行入场调研采样分析,本研究发现不同行业中VOCs组成特征有差异,多数行业VOCs物种排放以卤代烃、芳香烃、烷烃等为主,纺织印染业、制鞋业等部分行业以含氧有机物排放为主。通过调研和实测对部分行业的VOCs排放因子水平做了本地化深入研究,调研统计青岛市约49%的企业安装了VOCs治理设施;在企业所安装的VOCs治理设施中吸附法占比最大,占比为26%。