AHP与GA-SVM耦合模型在滑坡预警中的应用*

为了减少滑坡造成的损失,提高滑坡预测的准确性,通过搭建灾害模拟平台获得滑坡的实验数据,在获得多组模拟实验数据后,分析各变量的特性。首先,通过层次分析(Analytic Hierarchy Process,AHP)算法,对滑坡进行危险度划分;然后,通过支持向量机(Support Vector Machine,SVM)建立模型,遗传算法(Genetic Algorithm,GA)再优化SVM参数,提出1种层次分析法与GA-SVM相耦合的模型。研究结果表明:AHP方法划分后的数据,通过GA与SVM结合建立的模型精度较好,实验预测结果与实际结果较为吻合,与单一SVM相比,精度更高,结果更好,更加适用于多变量的复杂非线性滑坡预警。     

四川省眉山市“三线一单”编制案例分析

通过四川省眉山市“三线一单”编制试点工作的开展,探索出适合四川省情、尺度得当、具有可操作性的环境管控单元划分方法和适用技术方法以及生态环境准入清单制定原则,为全省“三线一单”编制工作的顺利推进起到了示范和引领作用。     

农村变电站生活污水处理现状与探讨

我国大部分变电站位于农村地区,因无接入城市污水管网条件,农村变电站生活污水需就地处理后达标排放。变电站生活污水存在水量少、时变化系数大等特点,现有处理工艺不能正常运行或无法达标排放。本文借鉴农村生活污水处理技术,提出了生物组合法处理对策,可为变电站生活污水处理提供参考。     

The influence on carbon,nitrogen recycling,and greenhouse gas emissions under different C/N ratios by black soldier fly

Environmental Science and Pollution Research - Currently, sustainable utilization, including recycling and valorization, is becoming increasingly popular in waste management. Black soldier fly...     

临安区不同功能区道路降雨径流重金属污染特征及源解析

于2017—2018年对临安区不同功能区的道路进行道路降雨径流采样,分析降雨量、季节变化和街尘对临安区道路降雨径流重金属污染特征的影响,并运用主成分分析法进行重金属源解析。结果表明,临安区道路降雨径流污染较严重;降雨强度越大,重金属浓度下降更快且末期浓度更低;道路降雨径流中Cu呈现春夏季大于秋冬季,Zn、Cr呈现秋冬季大于春夏季,Pb、Ni呈现冬春季大于夏秋季的季节变化趋势;临安区街尘与道路降雨径流重金属污染存在一定的相关性;主成分分析表明重金属污染源主要为以电线电缆为主的工业活动和汽车损耗。     

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...     

神农架大九湖后生浮游动物群落结构和水质评价

大九湖是南水北调中线工程的重要水源涵养地之一。为了解实施生态修复工程后浮游动物群落结构特征及水质污染情况,于2014年11月（枯水期）、2015年5月（平水期）和9月（丰水期）进行了3次采样调查。研究表明:共检出后生浮游动物36种,其中桡足类2种、枝角类5种和轮虫29种,轮虫丰度在3次采样中均占总丰度的80%以上,仅有的几种枝角类和桡足类也是小型种类或以幼体为主。11月优势种是寡污-β中污型-螺形龟甲轮虫（Keratella cochlearis）和晶囊轮虫（Asplachna sp.）;5月优势种为寡污-β中污和β中污型-螺形龟甲轮虫、迈氏三肢轮虫（Filinia maior）、无棘螺形龟甲轮虫（Keratella cochlearis tecta）、广生多肢轮虫（Polyarthra vulgaris）和萼花臂尾轮虫（Brachionus calyciflorus）;9月优势种是β-中污型-裂足臂尾轮虫（Brachionus diversicornis）、迈氏三肢轮虫和剪形臂尾轮虫（Brachionus forficula）。冗余分析（RDA）表明高锰酸盐指数(CODMn)是影响后生浮游动物污染类型分布的最关键环境因子。 关键词： 大九湖;后生浮游动物;污染类型;高山湿地湖泊     

Using Si depletion in aerosol to identify the sources of crustal dust in two Chinese megacities

Depletion of Si in transported dust has been recognized for many years. It can be used to distinguish between transported and local dust in cities, although it rarely has been. Here we use the variations of the Si/Al ratio in 15 months of continuous PM_2.5 samples at Beijing (northern China) and Chongqing (southwestern China) to reveal the seasonal patterns of their dust sources. For both cities, peaks of concentration for Si and Al in PM_2.5 corresponded with minima of Si/Al, and could often be linked to pulsed air flow from deserts to the northwest. With significant depletion (up to 80%) and homogeneous distribution at urban and rural sites, Si/Al showed a clear seasonal evolution, which decreased from spring to summer, increased from fall to winter, and collapsed during Chinese Spring Festival, indicating the dominance of transported dust, local fugitive dust and firework influence, respectively. The low ratios implied that desert dust is a common source during spring at Chongqing, whereas its presence during cold season at Beijing was also more frequent than expected. Failing to recognize the depletion of Si may lead to an overestimate of desert dust by 15%–65% when using the average abundance of Al in crust (6%–8%), as in previous studies. The difference in Si/Al ratio between local and transported dust implies that >60% of the dust at Beijing came from outside the city during the springs of 2004–2006. This result can help resolve the contradictory findings on this topic that have been presented earlier.