基于MCMC法的水质模型参数不确定性研究

参数识别是数学模型应用的前提.鉴于常用贝叶斯离散化方法在搜索复杂模型参数后验分布时的计算限制的原因,本文引入了MCMC采样法.为考察MCMC法对参数后验分布的搜索性能和效率,进行了2个案例研究.结果表明,MCMC法对参数后验分布的搜索,无论是搜索性能还是搜索效率,均表现出了独特的优越性.同时,Gelman收敛判别准则计算表明,MCMC采样序列均能稳定收敛到参数的后验分布上.可见,MCMC法适用于复杂环境模型的参数识别和不确定分析研究.     

大气甲烷碳同位素测试方法及其在雅克拉凝析气田上方大气中的应用

研制了一套与气体稳定同位素质谱仪联机在线分析大气甲烷碳同位素的制样技术,该方法具有精度较高(±0.4×10^-3)、用样量少(200mL)、耗时短(45min)和操作相对较为简便的优点.利用该系统对新疆塔里木盆地雅克拉凝析气田上方大气甲烷的碳同位素组成做了初步测试.结果表明,其大气甲烷碳同位素组成平均值为-45.0×10^-3,相对全球大气甲烷碳同位素平均值明显偏重1.2×10^-3～2.0×10^-3,显示了其甲烷主要来源为深部凝析气藏中的甲烷等气体的渗漏和扩散.并且由于白天日照强,温度高,气田渗漏和扩散到地表的甲烷部分被氧化,因此白天大气甲烷的浓度略低于夜晚,且¹³C_CH4较夜晚偏轻.     

Fuzzy相似优先比鉴别海面溢油

建立了海南溢油鉴别的Ｆｕｚｚｙ相似优先比模式,提出了油种鉴别的Ｆｕｚｚｙ相似优先比置信水平λ↑－的模糊区概念。海面溢油鉴别主要受油种、实验误差和风化的影响。任丘原油风化１个月同其它１３种油可鉴别的Ｆｕｚｚｙ相似优先比置信水平的范围为０．００５８１９≤λ↑－≤０．０４６４１,大庆原油在海面上风化２８ｄ同其它５种油可鉴别的范围为０．１１０１≤λ↑－≤０．２５１０。海面溢油鉴别结果表明,本法较指纹谱图辨     

火灾场景设计与火灾危险度分析在火灾性能化设计危险源辨识中的应用

通过介绍设定火灾场景,计算热释放速率,使用火灾危险度的概念对建筑火灾的危险性进行综合评价等方法,探讨了在性能化防火设计中的危险源辨识问题。给出了一个对某航站楼进行面向性能化设计的火灾危险评估的工程算例。     

Identification of the bound residue composition derived from 14C-labeled chlorsulfuron in soil by using LC-MS and isotope tracing method

A new method for extracting the bound residue(BR) derived from ^14C-iabeled chlorsulfuron in soils was developed, and the technique of combining LC-MS with isotope tracing method was subsequently applied to identify the composition of the ^14C-BR in a loamy Fluvent derived from manne deposit. The results showed that the ^14C- [ 2- amino-4- methoxyl-6- methyl- 1,3,5 ]-triazine, ^14C- [ 2-amino-4-hydroxyl-6-methyl-1,3,5]-tdazine and ^14C-chlorsulfuron parent compound constituted the main composition of the ^14C-BR denved from ^14C-labeled chlorsulfuron in the soil. The radioactive ratio of three compounds accounted for 39.8 %, 35.4 % and 17.9 % of total recovered radioactivity, respectively. However, a small amount(3.6 % of total recovered radioactivity) of the complex of ^14C- [ 2-amino-4-hydroxyl-6-methyl-1,3,5]-tdazine might have existed in the ^14C-BR in association with an unknown soil substrate. 2-chlorobenzenesulfonamide was also detected to be one of the components of the BR. The results could well explain the mechanism of phytotoxicity caused by the BR derived from chlorsulfuron in soil. In addition, the mechanism of BR formation in soil was also discussed in details.     

基于IPSO-BP神经网络的钢丝绳断丝损伤识别模型研究

为解决传统钢丝绳断丝损伤识别方法精度低,BP神经网络陷入局部最优等问题,提出改进粒子群算法（IPSO）的BP神经网络识别模型。通过采集钢丝绳断丝损伤信号,提取缺陷信号特征,用峰值、峰峰值、波宽、波形下面积和波动能量5个特征值组成特征向量作为神经网络的输人,断丝数量作为神经网络的输出;利用改进粒子群算法对BP神经网络的初始权值和阈值进行优化;建立基于IPSO-BP算法的神经网络模型,用于钢丝绳断丝的定量识别。结果表明:IPSO-BPS神经网络模型的钢丝绳断丝损伤识别精度、泛化能力均高于传统BP神经网络模型,且改进的粒子群算法迭代寻优速度更快。     

A meta-analytic test of the differential pathways linking ethical leadership to normative conduct

We test how ethical leadership influences normatively (in)appropriate work behavior through distinct mediating pathways, including one's exchange relationship with the leader, ethical culture, and identification with the organization. Our study also controls for transformational leadership as a predictor and trust in leader as a nonhypothesized alternative mechanism. We test our hypotheses using meta-analytic structural equation modeling based on our meta-analysis of 301 independent samples (N = 103,354) and relevant meta-analytic correlations reported in previous research. Supporting our prediction, we found that leader–member exchange, which represents social exchange theory, was the most potent mechanism that accounts for the positive relationship between ethical leadership and task performance. In contrast, ethical culture, which assesses a social learning mechanism, is the strongest predictor of counterproductive behavior. In addition, all three hypothesized mediators each contribute to understanding the positive relationship between ethical leadership and organizational citizenship behavior, although the indirect effect via organizational identification was the weakest. The findings hold after controlling for job satisfaction as another mediator parallel to the theoretical ones. Our results contribute to a precise theory about ethical leadership by differentiating the processes through which it affects employee behavior.     

生态环境监测工作的安全风险及防控措施

随着经济社会的发展,人们的生活水平不断提高,公众对生态环境质量的需求也在增加。生态环境监测作为生态环境保护的一项重要的基础性工作,是生态环境治理的数据支撑和重要保障,也是公众了解环境质量必不可少的途径。笔者结合2020年3月6日实施的《职业健康安全管理体系要求及使用指南》等规范要求,调研生态环境监测职业面临的安全风险,识别和分析环境监测工作可能存在的安全风险因素,并在此基础上给出相应的防护技术及措施,对生态环境监测从业人员的安全保障提出了重点抓好生态环境监测实验室的安全与质量管理、制定应急预案、强化日常监管、完善制度体系和加强宣传教育等方面的建议。     

对美国格林威治湾底栖动物造成危害的潜在胁迫因子的诊断

格林威治湾是纳拉甘西特湾一处受到多种影响因子作用的城市化形成港湾。本研究鉴定了对于格林威治湾底栖动物的重要影响因素。首先利用现存数据与信息来验证水体受损情况。其次确定了影响来源,影响因素和具体影响的存在。继而探究了来源,影响因素和具体影响之间的关系。这使我们能够鉴定出最可能危害水体的影响因素。本研究评估了化学制品、营养物质和悬浮沉积物3类污染物。这种证据权重的方法表明格林威治湾主要受到富营养化相关的影响因素危害。格林威治湾的沉积物富含碳,低溶解氧浓度的情况十分常见,尤其在格林威治湾西部。底栖生物群落出现发育不良的情况,与在氧气不足情况下的预测结果吻合。尽管我们的分析表明环境中的污染物浓度可能导致有害结果的出现,我们并未在结果中检出毒性。这是由于沉积物中存在大量的有机物,这些有机物限制了污染物的生物可利用性。我们的分析还表明悬浮沉积物的影响对于湾区的大部分区域而言几乎不存在。本分析展示了诊断法步骤可以用于组织和评估影响格林威治湾生态健康的潜在影响因素。诊断法步骤对于管理受到多重因素影响的水体非常有用。
精选自Marguerite Pelletier, Kay Ho, Mark Cantwell, Monique Perron, Kenneth Rocha, Robert M. Burgess, Roxanne Johnson, Kenneth Perez, John Cardin, Michael A. Charpentier. Diagnosis of potential stressors adversely affecting benthic invertebrate communities in Greenwich Bay, Rhode Island, USA. Environmental Toxicology and Chemistry: Volume 36, Issue 2, pages 449–462, July 2017. DOI: 10.1002/etc.3562
详情请见http://onlinelibrary.wiley.com/wol1/doi/10.1002/etc.3562/full
     

大气环境质量评价的属性识别法

阐述了评价大气环境质量的属性识别法,介绍了属性识别理论模型和属性权重的确定方法,并根据GB 3095—1996((大气环境质量标准》,分别采用模糊数学综合评价法和属性识别法对武汉市大气环境监测结果进行了实例评价,结果表明,用属性识别法评价的结果比用模糊数学综合评价法评价的结果更能反映真实情况。