极端梯度提升与随机森林融合的天然气露点预测方法

目的 解决目前水露点数据多为人工采用测量仪器测得,时效性低且成本高昂的问题。方法 建立一种基于极端梯度提升(XGBoost)和随机森林(RF)的天然气水露点预测方法。采用XGBoost方法对所有监测工艺参数进行分析,筛选出主要影响水露点的关键工艺特征参数,以排除无关特征参数对预测的干扰。建立RF预测模型,输入关键特征集参数,实现对水露点的实时预测。以重庆气矿某脱水监测系统监测数据与生产数据为例,对所提预测方法进行对比分析验证。结果 相较于XGBoost、SVM等预测方法,RF模型具有最佳的预测性能,且经过XGBoost特征选择后,RF预测结果的MAE值降低了0.016 9 ℃,RMSE值降低了0.014 6 ℃。结论 基于极端梯度提升与随机森林融合的水露点预测方法具有更优的预测精度与鲁棒性,对指导脱水现场生产具有积极作用。     

Molecularly imprinted polymer for analysis of trace atrazine herbicide in water

A molecularly imprinted polymer (MIP) for atrazine was synthesized by non-covalent method. The binding capacity of MIP was 1.00 mg g^{? 1} polymer. The selectivity and recovery were investigated with various pesticides which are mostly, found in the environment, for both similar and different chemical structure of atrazine. The competitive recognition between atrazine and structurally similar compounds was evaluated and it was found that the system provided highest recovery and selectivity for atrazine while low recovery and selectivity were obtained for the other compounds. The highest recovery was obtained from MIP compared with non-imprinted polymer (NIP), a commercial C₁₈ and a granular activated carbon (GAC) sorbent. The method provided high recoveries ranged from 94 to 99% at two spiked levels with relative standard deviations less than 2%. The lower detection limit of the method was 80 ng L^{? 1}. This method was successfully applied for analysis of environmental water samples.     

A Cumulative Energy Demand indicator (CED), life cycle based,for industrial waste management decision making

Life cycle thinking is a good approach to be used for environmental decision-support, although the complexity of the Life Cycle Assessment (LCA) studies sometimes prevents their wide use. The purpose of this paper is to show how LCA methodology can be simplified to be more useful for certain applications.In order to improve waste management in Catalonia (Spain), a Cumulative Energy Demand indicator (LCA-based) has been used to obtain four mathematical models to help the government in the decision of preventing or allowing a specific waste from going out of the borders. The conceptual equations and all the subsequent developments and assumptions made to obtain the simplified models are presented.One of the four models is discussed in detail, presenting the final simplified equation to be subsequently used by the government in decision making.The resulting model has been found to be scientifically robust, simple to implement and, above all, fulfilling its purpose: the limitation of waste transport out of Catalonia unless the waste recovery operations are significantly better and justify this transport.     

广州市“南肺”果园土壤H g和A s的形态分析

采用Tessier连续提取法、BCR连续提取法和Leleyter连续提取法,研究广州"南肺"果园土壤Hg和As的形态分布和生物可给性。结果表明,果园土壤Hg和As形态分布都以残渣态为主;3种方法的土壤生物可给性分析均表明,Hg和As生物可利用态的含量较少;生物不可利用态是主要形态,其次是潜在可利用态。     

沸石负载介孔TiO_2的制备及其可见光催化性能研究

以人工沸石为载体,采用溶胶-凝胶法,制备出了沸石负载介孔改性TiO2光催化剂。以亚甲基蓝为目标污染物,研究了不同掺杂金属及介孔薄膜制备方法对纳米TiO2可见光催化性能的影响。用扫描电镜(SEM)、紫外-可见吸收光谱(UV-vis)等手段研究了掺杂改性前后TiO2的变化。研究发现,N-Fe掺杂可使催化剂的光吸收谱带红移。当掺杂Fe的比为0.2%,焙烧温度为500℃时制备的TiO2效果最好,在可见光照射2 h后,亚甲基蓝的去除率可达90%。研究表明经十六烷基三甲基溴化铵(CTAB)作为模板剂修饰后的TiO2催化效果有明显提高,通过将经过萃取去除模板剂的共掺TiO2和直接高温焙烧的TiO2对比后发现,前者比后者降解亚甲基蓝的效果要好,原因在于,高温直接焙烧去除模板剂很容易使模板剂形成的骨架塌陷,减小了比表面积,从而影响了其光催化性能。     

OBIA与RF结合的龙口市土地利用信息提取方法

为提高中分辨率遥感影像解译精度,本文提出面向对象影像分析（Object Based Image Analysis,OBIA）与随机森林（Random Forest,RF）结合的土地利用信息提取方法。采用Landsat 8 OLI影像,针对不同地物特点,阈值分割和多尺度分割结合创建影像对象,规则集和分类器协同分类,基于Relief F算法分别对光谱特征、纹理特征及所有特征降维筛选特征子集,并与全部特征一起应用RF建模,对龙口市进行土地利用信息提取与比较。结果表明：OBIA与RF结合提取土地利用信息,基于Relief F算法筛选纹理特征,保留完整光谱、几何、空间关系特征构建RF模型,建模错分率为0.0958,分类总体精度和Kappa系数分别为89.37%和0.872,取得较理想结果。该方法可应用于中分辨率遥感影像土地利用信息提取。     

Combined molecular mechanics, molecular dynamics and quantum mechanical study of (+)-multifidene structure and conformation

Molecular mechanics, molecular dynamics and quantum mechanical tools have been used to explore the conformational space of (+)(3S,4S)-multifidene bioactive compounds. All the computations have been performed in the absence and presence of the solvent, simulated by the explicit introduction of a number of water molecules or by the polarized continuum model (PCM). Three conformational minima, both in vacuo and in water (simulated by PCM tool), have been confirmed at ab initio level of theory (B3LYP and MP2). All minima lie in a very narrow range of energy, and the interconversion between them is possible with a small amount of energy expense.     

Broad range analysis of endocrine disruptors and pharmaceuticals using gas chromatography and liquid chromatography tandem mass spectrometry

Endocrine disrupting compounds (EDCs) and pharmaceuticals and personal care products (PPCPs) have been globally detected in impacted natural waters. The detection of trace quantities of EDCs and PPCPs in the environment is of great concern since some of these compounds have known physiological responses at low concentrations. EDCs can have a wide range of polarities, acidic and basic moieties, and exist in trace quantities, which often requires numerous complex extractions, large sample collection volumes, and multiple instrumental analyses. A comprehensive method has been developed allowing for the analysis of 58 potential EDCs in various water matrices using a single solid-phase extraction (SPE) of a 1 L sample with subsequent analyses using both gas chromatography and liquid chromatography, each coupled with tandem mass spectrometry (GC–MS/MS and LC–MS/MS). Instrument detection limits ranged between 0.12–7.5 pg with corresponding method reporting limits of 1–10 ng l⁻¹ in water. Recoveries for most compounds were between 50% and 112% with good reproducibility (RSD 6–22%).     

简化连续提取法评价污染土壤中Zn、Cd的植物有效性

以0.01 mol·L^-1CaCl₂和0.005 mol·L^-1 DTPA作为提取剂,用简化的3步连续提取法对贵州省赫章县土法炼锌污染土壤中Zn、Cd的形态进行了分析.结果显示,污染土壤中Zn、Cd主要以残渣态的形式存在,CaCl₂提取态和DTPA提取态Zn、Cd平均仅占全量的0.63%、3.91%和10.94%、10.13%.土壤中不同形态Zn、Cd含量与玉米中Zn、Cd含量的相关分析结果显示,CaCl₂提取态金属与玉米中金属含量没有显著的相关关系,而DTPA提取态、残渣态以及总量Zn、Cd与玉米根、茎叶中Zn、Cd含量显著正相关.这些结果表明CaCl₂提取态Zn、Cd对土壤中该元素的植物有效态可能不具重要贡献,而DTPA提取态金属和金属总量在一定程度上能作为评价土壤中元素植物有效性的标准.     

加速溶剂萃取-在线净化-液相色谱串联质谱法检测污泥堆肥样品中三氯生和三氯卡班

污泥堆肥样品基质复杂,其中三氯生（TCS）和三氯卡班（TCC）的提取和净化过程操作烦琐、回收率低。通过对净化条件进行优化,建立了加速溶剂萃取-在线净化-液相色谱串联质谱法检测污泥堆肥样品中TCS和TCC的分析方法,并对上海某污水厂实际污泥堆肥样品中的TCS和TCC进行定量分析。结果表明,TCS和TCC的检出限分别为0.003和0.0002μg/g,加标回收率为90.3%~99.8%,相对标准偏差（RSD）为1.2%~8.5%（n=6）。该方法样品前处理时间短、回收率高、试剂使用量少,能够满足实际污泥堆肥样品中TCS和TCC的检测要求。