基于遗传-神经网络的电网流域面雨量预报方法研究

以重点水力发电厂和大中型水库为主要考量,并兼顾地形地貌和中小河流的分布特征,将广西划分为23个电网流域,研究了基于非线性的神经网络电网流域面雨量预报方法。以5-6月龙滩近库区、龙江流域等6个电网流域为例,利用遗传算法优化BP神经网络的连接权和网络结构,建立了各电网流域的遗传-神经网络电网流域面雨量预报模型。对独立样本的预报结果表明,基于遗传-神经网络的电网流域面雨量预报模型的预报能力要优于传统的逐步回归预报模型,也明显优于日本、德国数值模式预报产品所换算成的电网流域面雨量预报,并与气象部门同期制作的综合面雨量预报产品能力相当,因而,遗传-神经网络面雨量集合预报模型有较好的业务应用前景。     

Bio-VOCTM呼吸采样器样品转移流程评价及优化

人体呼出气中挥发性有机化合物（VOCs）分析已应用于环境暴露评估和疾病诊断等领域,其中肺泡气可以排除来自周围环境以及口腔和鼻腔内死腔气体的干扰,准确反映暴露及内源性代谢物水平。对采集肺泡气的Bio-VOC^TM呼吸采样器转移样品至吸附管的常规操作流程进行评价及优化。使用Bio-VOC^TM呼吸采样器采集含有丙酮、异戊二烯、乙醇、正丁醇、正己烷和正庚烷6种VOCs的呼出气模拟样品,比较不同方式转移至吸附管对模拟样品转移效率的影响,采用热脱附-气相色谱-质谱法（TD-GC-MS）对吸附管中样品进行分析,并将优化转移方式应用于实际人群呼出气样品分析。结果表明：按照Bio-VOC^TM呼吸采样器常规转移方式,模拟样品中6种VOCs的转移效率均低于50%且不稳定;降低转移流量后,转移效率有所提高,进一步优化后转移效率可达90.0%—101.0%,相对标准偏差均小于5%。转移方式优化后,检测到的人体呼出气样品中3种VOCs的峰面积响应值显著提高（P＜0.01）。优化方式提高了使用Bio-VOC^TM呼吸采样器采集肺泡气进行VOCs分析的准确性和可靠性,为肺泡气采集方法的标准化提供了科学依据。     

Low-reactive circulating fluidized bed combustion (CFBC) fly ashes as source material for geopolymer synthesis

In this contribution, low-reactive circulating fluidized bed combustion (CFBC) fly ashes (CFAs) have firstly been utilized as a source material for geopolymer synthesis. An alkali fusion process was employed to promote the dissolution of Si and Al species from the CFAs, and thus to enhance the reactivity of the ashes. A high-reactive metakaolin (MK) was also used to consume the excess alkali needed for the fusion. Reactivities of the CFAs and MK were examined by a series of dissolution tests in sodium hydroxide solutions. Geopolymer samples were prepared by alkali activation of the source materials using a sodium silicate solution as the activator. The synthesized products were characterized by mechanical testing, scanning electron microscopy (SEM), X-ray diffractography (XRD), as well as Fourier transform infrared spectroscopy (FTIR). The results of this study indicate that, via enhancing the reactivity by alkali fusion and balancing the Na/Al ratio by additional aluminosilicate source, low-reactive CFAs could also be recycled as an alternative source material for geopolymer production.     

Comprehensive and high-resolution emission inventory of atmospheric pollutants for the northernmost cities agglomeration of Harbin-Changchun, China: Implications for local atmospheric environment management

Using a bottom-up estimation method, a comprehensive, high-resolution emission inventory of gaseous and particulate atmospheric pollutants for multiple anthropogenic sectors with typical local sources has been developed for the Harbin-Changchun city agglomeration (HCA). The annual emissions for CO, NOx, SO₂, NH₃, VOC_S, PM_2.5, PM₁₀, BC and OC during 2017 in the HCA were estimated to be 5.82 Tg, 0.70 Tg, 0.34 Tg, 0.75 Tg, 0.81Tg, 0.67 Tg, 1.59 Tg, 0.12 Tg and 0.26 Tg, respectively. For PM₁₀ and SO₂, the emissions from industry processes were the dominant contributors representing 54.7% and 49.5%, respectively, of the total emissions, while 95.3% and 44.5% of the total NH₃ and NOx emissions, respectively, were from or associated with agricultural activities and transportation. Spatiotemporal distributions showed that most emissions (except NH₃) occurred in November to March and were concentrated in the central cities of Changchun and Harbin and the surrounding cities. Open burning of straw made an important contribution to PM_2.5 in the central regions of the northeastern plain during autumn and spring, while domestic coal combustion for heating purposes was significant with respect to SO₂ and PM_2.5 emissions during autumn and winter. Furthermore, based on Principal Component Analysis and Multivariable Linear Regression model, air temperature, relative humidity, electricity and energy consumption, and the urban and rural population were optimized to be representative indicators for rapidly assessing the magnitude of regional atmospheric pollutants in the HCA. Such indicators and equations were demonstrated to be useful for local atmospheric environment management.     

地冲击荷载作用下双层结构磁流变阻尼器隔震效果的数值模拟

在抗爆结构中,一般采用单层隔震系统进行隔震,目前单层隔震系统的最大隔震率可以达到90%以上,但在加速度峰值很高的冲击荷载作用下,隔震后的结构响应加速度仍然很大。鉴于这种情况,本文对加入磁流变阻尼器(MRD)的双层隔震系统进行了研究。针对抗爆结构的两种典型荷载,采用改进Bouc-Wen模型和模糊控制方法,利用Matlab Simulink对双层隔震系统进行了数值模拟,计算了不同荷载作用下,不同隔震系统的加速度、位移及结构的振动剂量值(VDV)的响应,并与结构采用单层隔震系统的结果进行了对比。结果表明,与单层隔震系统相比,带有磁流变阻尼器的双层隔震系统没有太多的优越性。     

Heterogeneous Fenton degradation of ofloxacin catalyzed by magnetic nanostructured MnFe2O4 with different morphologies

Environmental Science and Pollution Research - Magnetic nanostructured MnFe2O4 with different morphologies, synthesized via chemical co-precipitation and hydrothermal method, was assayed as...     

微生物对汞矿区农田土壤汞甲基化的影响

为研究典型汞矿区农田土壤中汞的甲基化作用,实验以受汞污染的旱田土壤和稻田土壤为对象,分别进行灭菌,促进硫酸盐还原菌(SRB)活性处理,抑制SRB活性处理以及促进铁还原菌(FeRB)活性处理,分析非微生物作用和微生物作用对土壤甲基汞(MeHg)生成的影响.结果表明,促进SRB活性处理的土壤MeHg生成量最高,其中旱田土MeHg增量为0.15~0.38μg·kg~(-1),稻田土壤MeHg增量为1~2μg·kg~(-1);抑制SRB活性处理和促进FeRB活性处理的MeHg增量较小,最高值仅为0.025μg·kg~(-1).相比于旱田土壤,稻田土壤具有更高的汞甲基化能力,其MeHg生成量是旱田土壤的4~9倍.土壤SRB数量与MeHg生成量具有相同的变化趋势,二者具有显著的正相关性(R~2=0.57,P0.01).因此,该研究区土壤汞甲基化作用主要是微生物汞甲基化作用,且主要的汞甲基化细菌是SRB.此外,稻田是农田中MeHg生成的活跃地区,在评估和控制MeHg对人体健康危害时需要重点关注.     

黑龙江及周边中强震前不同参数模型矩加速释放(AMR)现象特征分析

应用时间破裂分析方程中表征地震距释放程度的m值,选取不同空间尺度进行时空扫描研究,得到震前1年以每2个月为时间间隔的空间m值随时间变化的分布图像,运用不同参数设置计算了地震前Benioff累积应变释放特征。通过对比这些变化找出研究区内m值空间分布搜索主震破裂点扫描效果较好的参数模型,尝试性地给出了"矩释放程度空间扫描"方法应用于地震预测的经验指标。并结合与矩加速释放具有相同物理机制的加卸载响应比(Load/Unload Response Ratio)方法进行综合分析,利用这两种方法对地震进行中短期预测,达到两种方法互相检验、相互辅助的目的。考虑到本文仅对6个震例进行研究,无法给出具有统计显著性的结论,所以仍需结合其他方法进行点、线、面的综合分析。     

Electrochemical and spectroscopic characteristics of dissolved organic matter in a forest soil profile

Dissolved organic matter （DOM） represents one of the most mobile and reactive organic compounds in ecosystem and plays an important role in the fate and transport of soil organic pollutants, nutrient cycling and more importantly global climate change. Electrochemical methods were first employed to evaluate DOM redox properties, and spectroscopic approaches were utilized to obtain information concerning its composition and structure. DOM was extracted from a forest soil profile with five horizons. Differential pulse voltammetry indicated that there were more redox-active moieties in the DOM from upper horizons than in that from lower horizons. Cyclic voltammetry further showed that these moieties were reversible in electron transfer. Chronoamperometry was employed to quantify the electron transfer capacity of DOM, including electron acceptor capacity and electron donor capacity, both of which decreased sharply with increasing depth. FT-IR, UV-Vis and fluorescence spectra results suggested that DOM from the upper horizons was enriched with aromatic and humic structures while that from the lower horizons was rich in aliphatic carbon, which supported the findings obtained by electrochemical approaches. Electrochemical approaches combined with spectroscopic methods were applied to evaluate the characteristics of DOM extracted along a forest soil profile. The electrochemical properties of DOM, which can be rapidly and simply obtained, provide insight into the migration and transformation of DOM along a soil profile and will aid in better understanding of the biogeochemical role of DOM in natural environments.