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。     

海洋巨藻生物吸附剂对Hg~(2 )吸附性能的研究

就丰产廉价的海洋巨藻(Durvillaea Potatorum)所制的生物吸附剂对Hg2 的吸附性能进行了研究,该生物吸附剂经特殊的稳定化预处理而得。研究表明,该生物吸附剂的Hg2 吸附容量对溶液的pH值十分敏感,在pH=3时,最大吸附容量达到3.1mmol/g(干重),在溶液pH=0.4～5.0的范围内,Hg2 的吸附容量为0.49—3.1mmol/g(干重);溶液中的C1-能显著地降低生物吸附剂对Hg2 的吸附容量;其对Hg2 的去除率可达到99％以上。该生物吸附剂可以有效地去除工业废水中的Hg2 。     

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。     

地铁车站突发客流疏运能力的理论计算与分析

针对车站OB应对亚运会期间大规模突发客流疏导问题,利用理论分析的方法计算不同车站设计方案,包括车站一岛一侧站台设计和一岛站台设计及不同通道设计时的最大疏运能力。计算考虑了车站建筑结构、出入口、闸机口和边门、楼扶梯等通道的数量及通过能力、站台面的容纳能力、列车的装载量、售检票方式、行车交路组织等诸多因素。通过对影响客流疏导的主要控制因素进行能力核算,分别得到了两种站台形式设计方案时的最大疏运能力的理论值。研究结论将有利于解决该车站应对亚运会期间大规模突发客流疏导问题,同时理论分析方法可为其他类似的车站能力设计提供参考。     

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.     

Pesticide Impact Rating Index – A Pesticide Risk Indicator for Water Quality

Pesticide users, natural resource managers, regulators, government agencies and many others are concerned about the off-site impact of pesticides on the environment. Systematic methods of the assessment of potential risk of pesticides to environmental components can serve as valuable tools in decision making and policy formulation. Simple risk indicators have been developed which cover a range of scenarios such as toxicity to organisms, health of farm workers, consumer health, and residues in harvested produce. The authors have developed a software package named Pesticide Impact Rating Index (PIRI) that outputs an improved pesticide risk indicator for water quality. PIRI is a standalone, user-friendly, platform-independent program. It can be used to (i) rank pesticides in terms of their relative pollution potential to groundwater or surface water, and (ii) to compare different land uses in a catchment or at a regional scale in terms of their relative impact on water quality. It is based on pesticides use; the pathway through which the pesticides are released to the water resources (drift, runoff, erosion, leaching) and the value of the water resources threatened. Each component is quantified using pesticide characteristics (including toxicity to organisms at different trophic levels, i.e. fish, daphnia, algae, etc.), environmental and site conditions (e.g. organic carbon content of soil, water input, slope of land, soil loss, recharge rate, depth of water table, etc.). This paper describes two case studies of the application of PIRI in Australia. The comparison of the risk assessment by PIRI on these revealed that PIRI correctly estimated the pollution potential of pesticides in greater than 80% of cases. A GIS version of PIRI is described in a companion paper in this volume. An erratum to this article is available at .     

Plant blindness and the implications for plant conservation

Plant conservation initiatives lag behind and receive considerably less funding than animal conservation projects. We explored a potential reason for this bias: a tendency among humans to neither notice nor value plants in the environment. Experimental research and surveys have demonstrated higher preference for, superior recall of, and better visual detection of animals compared with plants. This bias has been attributed to perceptual factors such as lack of motion by plants and the tendency of plants to visually blend together but also to cultural factors such as a greater focus on animals in formal biological education. In contrast, ethnographic research reveals that many social groups have strong bonds with plants, including nonhierarchical kinship relationships. We argue that plant blindness is common, but not inevitable. If immersed in a plant‐affiliated culture, the individual will experience language and practices that enhance capacity to detect, recall, and value plants, something less likely to occur in zoocentric societies. Therefore, conservation programs can contribute to reducing this bias. We considered strategies that might reduce this bias and encourage plant conservation behavior. Psychological research demonstrates that people are more likely to support conservation of species that have human‐like characteristics and that support for conservation can be increased by encouraging people to practice empathy and anthropomorphism of nonhuman species. We argue that support for plant conservation may be garnered through strategies that promote identification and empathy with plants.     

ABC算法优化SVR的磨损故障预测模型

目的为了提高故障预测的精度,针对支持向量回归SVR(Support vector machine for regression,SVR)参数选择困难的问题,提出一种采用人工蜂群(artificial bee colony,ABC)算法优化支持向量回归(SVR)的故障预测模型(ABC-SVR)。方法该模型先对样本数据进行重构,然后将故障预测误差(适应度)作为优化目标,通过ABC算法寻优找到最优的SVR参数,建立故障预测模型。最后通过实例仿真验证模型的优越性。结果采用ABC算法优化的SVR故障预测模型进行时间序列预测,能够较好地跟踪发动机滑油金属元素浓度的变化过程,并且能够提前2个取样时间预测异常情况的出现。结论 ABC-SVR模型有效解决了SVR参数选择难题,能够更加准确地表现故障变化规律,提高了故障预测精度。