基于SOM-SVM组合分类器的涂层防护性能研究

目的为避免EIS,EN技术可能出现的问题,建立一个准确、高效的评价模型,以探究现役军用有机涂层防护性能。方法利用电化学阻抗谱(EIS)、电化学噪声(EN)技术分析了两种军车有机涂层在循环暴露试验中的腐蚀行为,提取低频阻抗模值|Z|_(0.1 Hz)与涂层噪声电阻R_n两种电化学评价参数作为自组织神经网络(SOM)的输入训练样本,同时结合支持向量机(SVM)方法建立涂层防护性能组合分类器。结果将涂层失效过程自适应地分为涂层防护性能良好、防护性能下降、基本失效三个阶段。结论所建立的SOM-SVM组合分类器对于辅助分析涂层防护性能具有可行性。     

Modeling the spatial pattern of farmland using GIS and multiple logistic regression: a case study of Maotiao River Basin, Guizhou Province, China

Land use change is an important topic in the field of global environmental change and sustainable development. Land use change modeling has attracted substantial attention because it helps researchers understand the mechanisms of land use change and assists regulatory bodies in formulating relevant policies. Maotiao River Basin is located in the province of Guizhou, China, which has a developed agricultural industry in the karst mountain areas. This paper selected biophysical and social–economic factors as independent variables, and constructed a multiple logistic regression of farmland spatial distribution probability by random sampling. Then, by using GIS technology and integrating the 2000 data, this study predicted the farmland spatial pattern. When the predicted map was compared with the actual farmland map for 2000, we noted that 71% of the simulation is in accordance with the 2000 farmland pattern. The result satisfactorily proves the reasonability and applicability of our model.     

Using environmental stressor information to predict the ecological status of Maryland non-tidal streams as measured by biological indicators

In Maryland, U.S., an interim framework has recentlybeen developed for using biologically based thresholds, or `biocriteria', to assess the health of nontidal streams statewide at watershed scales. The evaluation of impairment is based on indices of biological integrity from the Maryland Biological Stream Survey (MBSS). We applied logistic regression to quantify how the biotic integrity of streams at a local scale is affected by cumulative effects resulting from catchment land uses, point sources, and nearby transmission line rights-of-way. Indicators for land use were developed from the remote sensing National Land Cover Data and applied at different scales. We determined that the risk of local impairment in nontidal streams rapidly increases with increased urban land use in the catchment area. The average likelihood of failing biocriteria doubled with every 10% points increment in urban land, thus an increase in urban land use from 0 to 20% quadruples the risk of impairment. For the basins evaluated in this study, catchments with more than 40–50% urban land use had greater than 80% probability of failing biocriteria, on average. Inclusion of rights-of-way and point sources in the model did not significantly improve the fit for this data set, most likely because of their low numbers. The overall results indicate that our predictive modeling approach can help pinpoint stream ecosystems experiencing or vulnerable to degradation.     

常压塔塔顶腐蚀关键参量相关性分析与预测

预测常压塔顶回流罐切水的关键离子浓度可以为常压塔顶系统工艺防腐提供技术指导。收集了某炼厂2014-2016年常压装置的生产实时工艺参数、原料及产品采样数据、水质分析化验参数,采用线性插值方法对不同时间尺度的数据进行补全。通过线性相关性分析获得了影响常压塔顶回流罐切水总铁离子浓度、pH的主要因素,并基于深度学习、支持向量机回归、粒子群优化方法建立了腐蚀关键参量预测模型。结果表明,pH值和总铁离子浓度的相关因素大部分重叠,与原料性质及产品馏出温度较强相关;建立的回归模型预测精度高,在训练集和预测集上,总铁离子浓度预测值与测量值最大偏差分别为4. 4%和9. 8%,pH值预测值与测量值最大偏差分别为0. 9%和1. 1%。     

用VB语言编制水环境管理决策支持系统

在分析国内外环境管理决策支持系统（ＥＤＳＳ）开发现状的基础上，提出了采用Ｍｉｃｒｏｓｏｆｔ Ｖｉｓｕａｌ Ｂａｓｉｃ来开发水环境管理决策支持系统（ＷＥＭＤＳＳ）的方法，并利用程序语言编制了广东省潭江的ＷＥＭＤＳＳ。实例研究表明，充分利用ＶＢ编程语言的特点，可以快速方便地完成ＷＥＭＤＳＳ的结构设计和实现系统的各项功能。     

网络计划在液化气球罐内外部检验过程中的应用

在以工期考虑为重点的V-3002B液化气球罐内外部检验过程中，引用现代管理中网络计划技术关键路线法，为加快芳烃厂压力容器检验提供了科学的依据。本文总结和介绍了网络计划技术在芳烃厂贮运车间V-300亚球罐内外部检验过程应用情况。     

支持向量机用于芳烃类化合物对芳烃受体亲和性QSAR研究

尝试将支持向量机(SVM)应用于3种典型芳烃类环境毒物(PCDD,PCDF和PCB)定量构效关系研究,通过对芳烃受体亲和性考察,结果发现该组样本的生物活性在一定程度上与分子电性距离矢量具有非线性联系.SVM对内部和外部样本都具良好稳定性能和预测能力:所得模型拟合、交叉检验、外部预测复相关系数及均方根误差分别为R2cum=0.922、Q2cum=0.825、Q2ext=0.834和RMSext=0.531将其与文献报道及多元线性回归、偏最小二乘、人工神经网络进行比较,结果表明对小样本、非线性问题SVM具较强拓展性及泛化能力,故在环境毒物评价和控制中具有广阔应用前景.     

工业废水的生化耗氧过程Logistic动力学研究

在Monod方程的基础上推导了适用于描述工业废水的生化耗氧过程Logistic动力学模型,比较了Logistic动力学模型和一级动力学模型的区别。测定了废水的BOD5/CODcr,在此基础上对可生化废水的生化耗氧曲线进行分析,利用Logistic动力学模型对生化耗氧过程进行动力学研究,并比较各种不同稀释倍数时的kLog和S0。研究结果表明:随着稀释比的增大,耗氧过程的动力学参数kLog和计量系数S0增大,且S0和废水的BOD5非常接近。     

滑坡灾害危险性评价模型比较

通过对庆元地区滑坡灾害的详细研究，比较讨论了单变量评价、决策树评价与逻辑回归法等3种滑坡危险性评价模型在实际应用中的效果。单变量模型最为简单，计算方便，得到的结果也具有较高的准确性，当收集到的滑坡因素越多，对滑坡发育的刻画越精确，则最终评价结果也越可信；决策树算法能迅速从大量样本中归纳出一般规则，并具有较高的精度，但其结果也受到训练样本数量大小和样本维数即参与评价的因素集数量的影响；逻辑回归模型具有计算方法简单、评价结果物理意义明确等突出优点，但需要有大量样本数据的支持才能取得较好的效果。     

政策干预对提高煤矿职工素质影响探讨

煤矿井下作业人员素质低,是煤矿事故多发的重要原因之一.建议从大幅度提高煤矿井下人员工资待遇,吸引高素质人才从事煤矿工作;全方位开展煤矿职业培训工作,坚持先培训后就业,实行煤矿从业准入制;对现在岗的煤矿从业人员分期分批进行培训,实现煤矿从业人员持证上岗;政府出台政策,强制煤矿企业开展职工岗位技术培训和安全培训等4个方面着手,切实提高煤矿从业人员的整体素质,减少煤矿事故的发生.