Polybrominated diphenyl ether (PBDE) in blood from children (age 9–12) in Taizhou, China

Polybrominated diphenyl ethers (PBDEs) as ubiquitous persistent organic pollutants have attracted much attention in recent years. Exposure to PBDEs could induce a high health risk for children. The aim of this study was to investigate the PBDEs exposure of children (9–12 years) from Taizhou, China. Fifty-eight blood samples were collected in one school in a mountainous area in Taizhou. The concentrations of Σ9PBDEs (sum of BDE-28, -47, -99, -100, -153, -154, -183, -197 and -209) ranged from 2.66 to 33.9 ng/g lipid wet (lw) with a median of 7.22 ng/g lw. These concentrations were lower than those of children in USA, but close to European and Asian general population levels. The results showed that children in Taizhou countryside were at a low level of PBDEs exposure. The predominant congener was BDE-209, followed by BDE-28, -47, -197 and -153. High abundance of BDE-209 was consistent with the pollution background of PBDEs in China characterized by high brominated congeners as main pollutants.     

北方城市怎样发展低碳建筑？-以大连为调查样本

发展低碳建筑是一个系统性工程,涉及建造、运行使用、废弃等各个环节的节能减碳。受限于资金、技术等因素的制约,不可能一步到位,只能从“易”到“难”,分阶段、分步骤地推进。     

Validation of a conductometric bienzyme biosensor for the detection of proteins as marker of organic matter in river samples

This article describes a conductometric bi-layer based bienzyme biosensor for the detection of proteins as a marker of organic matter in rivers. Proteins were chosen to be used as indicators of urban pollution. The working mechanism of the bienzyme biosensor is based on the enzymatic hydrolysis of proteins into several fractions (peptides and amino acids), which results in a local conductivity change depending of the concentration of proteins. In this work, we began with the optimization of biosensor response using bovine serum albumin (BSA) as standard protein. For this objective seven enzymatic biosensors were prepared: four enzymatic sensors with only one layer of enzyme (proteinase K, trypsin, pronase or protease X) and three other enzymatic sensors with two layers (first layer: membrane containing proteinase K; second layer: one of the three other enzymes: trypsin, pronase or protease X). The biosensors were obtained through the deposition of enzymatic layers and the cross-linking process between enzymes and BSA in saturated glutaraldehyde vapour. The response of the various biosensors, described previously, were compared with the values of total organic carbon (TOC), and those of organic nitrogen (Norg), as determined by the laboratory accredits (CEMAGREF of Lyon) using the traditional method of analysis (NF EN 1484, infrared spectroscopy) and (NF EN 25663, mineralization/colorimetry assay) respectively for each water sample obtained from di erent sites in Lyon (France). The linear correlations obtained with the response of the seven biosensors showed the most important indices of correlations for the biosensor with two enzymatic layers: proteinase K + pronase (pkp). The optimum conditions for the preparation of the pkp biosensor increased the sensitivity and gave a limit of quantification of 0.583 g/L for TOC and 0.218 g/L for Norg in water samples. This sensor shows good reproducibility (2.28%), a capacity to be used at temperatures range 10– 30°C (depending on the season) and moreover a long lifetime (5 weeks).     

基于SPSS软件的居民住宅室内甲醛污染的调查与分析

甲醛是一种无色、具有强烈刺激性气味的气体,也是室内环境的主要污染物之一,人体长期暴露于高浓度甲醛环境中,会出现建筑病态综合症,如头晕、恶心、耳鸣等.目前学者对室内甲醛深度的影响因素、释放因子及其评价方法做了大量研究,但采用系统的统计学方法对甲醛含量及其影响因素进行分析尚且少见.     

基于多尺度动力学的空气质量指数特征分析

为了从不同尺度上挖掘空气质量指数及空气污染物浓度时间序列的动力特征,文章提出了一种多尺度趋势样本熵,该方法能有效处理具有强烈趋势的非线性时间序列。运用该方法研究了湖南省14个市州的AQI及PM_(2.5)和SO_2的动力学结构,发现AQI与PM_(2.5)的样本熵最接近,表明PM_(2.5)序列与AQI的演变趋势相似。此外,发现14个市州的3个时间序列在5～9 d的尺度上呈现最大的样本熵,表明以5～9 d为周期考虑的序列动力结构隐藏的信息量最大。进一步利用熵值对14个市州春夏秋冬4个季节进行了聚类,为寻找共同的污染源提供了依据,并为这些城市提出了空气污染防治的可行建议。     

融合多源异构数据的滑坡变形阶段智能判识方法

针对滑坡体不同变形阶段的监测数据样本不均衡,样本扩充量的限定研究较少以及判识模型准确率较低等现实问题,该文提出了一种少数类样本全局扩充量测算方法以及将分类结果混淆矩阵与GSA相结合的基于遗传的多分类样本合成方法MCGSA,可避免产生大量的合成样本,且有效解决了样本不均衡问题;其次借助堆栈泛化思想以及具有较强知识挖掘能力的机器学习模型,结合滑坡体的多源异构监测数据,构建了基于stacking的滑坡变形阶段智能判识模型;最后将该模型应用在多个滑坡隐患点上进行现场实验测试,并进行了对比实验分析,分析结果显示该判识模型的准确率可达89%,F1宏平均值达到了74%。模型的判识结果可为区域内滑坡隐患点的预警信息发布提供辅助决策。     

基于GWO-VMD算法的齿轮故障自适应特征提取

目的 齿轮产生故障时,利用其声发射信号进行自适应特征提取后诊断。方法 利用变分模态分解方法（VMD）对齿轮发生故障时的声发射信号进行分解。在现实状况中,采集声发射原信号噪声干扰大,导致特征提取准确度低,并且模态分解时参数需要人为调试设定。鉴于此,引入灰狼优化算法（GWO）,对模态分解个数k和二次惩罚因子α自适应选择最优参数后,对信号分解得到本征模态函数（IMF）。通过相关系数选出最佳IMF作为特征分量,计算其峭度和样本熵。结果 计算了各分量的相关系数,选取与原始信号最为相近的分量,分别计算其峭度和样本熵。分解后,齿轮故障声发射信号峭度高于正常的情况,而样本熵则偶然性表现为正常情况下的值大于故障条件下的值。结论 采用支持向量机对特征向量集进行分类识别,对比改进后的试验结果,GWO-VMD结合峭度–样本熵的方法能够有效地提取故障特征,判断齿轮状态是否健康。     

应急管理的大连样本

安全生产应急平台体系建设是安全生产"三项建设"的重要内容之一,是安全生产应急管理工作的重要保障。近年来,各级安全生产应急平台体系建设取得明显进展,部分地方应急平台建设已初具规模,在应急管理和应急协调指挥工作中发挥了重要作用。     

基于CEEMDAN与样本熵的高压断路器机械故障诊断

依据高压断路器振动信号特性，提出一种自适应白噪声完整经验模态分解( complete ensemble empirical mode decomposition with adaptive noise，CEEMDAN)与样本熵相结合的高压断路器故障特征提取方法。首先利用CEEMDAN将分闸振动信号分解成一系列内禀模态函数(intrinsic mode function，IMF)，然后利用相关系数法与归一化能量筛选包含信号主要特征信息的前7阶IMF分量，求取其样本熵作为特征量，最后采用粒子群算法(particle swarm optimization，PSO)优化支持向量机(support vector machine，SVM)分类器，对断路器不同故障类型进行分类识别。实验结果表明该特征提取方法能准确提取振动信号特征量，输入PSO-SVM诊断高压断路器故障能取得良好的效果。     

基于GAN的小样本腐蚀失厚率数据增强方法

目的 对小样本腐蚀失厚率数据进行数据增强,实现数据扩充,以提升后续分析模型的预测精度,减轻过拟合程度,并提升模型的泛化能力。方法 利用生成对抗网络(Generative Adversarial Networks,GAN)扩充腐蚀失厚率数据,使数据分布更加全面。对生成数据进行降维可视化分析,探究生成数据与原始数据样本的分布规律,分析数据增强合理性,并从多个算法模型、多个评价指标角度对分析预测能力、泛化能力进行评估。结果 生成数据填补了原始数据在样本空间分布的薄弱环节,加入生成数据后,各机器学习算法模型得出的MSE均值为未加入生成数据的61.72%~91.74%,皮尔逊均值为99.01%~113.64%,预测准确度提升,结果关联性更强,模型泛化能力增强。结论 GAN能有效对小样本腐蚀失厚率数据进行增强,数据扩充对分析预测提供正向支持,生成数据不宜多于原始数据,防止扰乱训练样本分布,同时存在生成数据多样性受限的问题。