基于GA-ELM浆体管道输送临界流速预测模型研究

针对浆体管道输送临界流速预测难度大、精确度低等技术难题,提出了基于极限学习机(ELM)的临界流速预测模型,用训练集对模型进行训练,以验证集预测值的均方误差作为适应度函数,利用遗传算法(GA)对ELM模型参数进行优化,应用优化得到的ELM模型对预测集进行预测。以某矿山为例,模型参数优化结果如下:隐含层节点数L为400,输入权值ai、偏置向量bi最优组合下预测结果适应度为0.0201。采用优化的ELM模型对预测集进行预测,预测结果的最大相对误差x=3.96%,平均相对误差y=1.58%,对比BP神经网络(x=12.95%)和SVM模型(x=3.19%),表明ELM模型更加精确、高效。     

气固两相流下球阀磨损特性研究

旋塞球阀是钻柱内防喷系统中的关键设备,在气固两相流下球阀易受磨损而失效,并造成严重的井喷事故。为此,将计算流体动力学理论与冲蚀磨损理论相结合,运用FLUENT软件对球阀壁面在气固两相流下的磨损分布情况进行研究,并进一步分析了球阀结构参数对于球阀壁面磨损的影响规律。结果表明:当气固两相流流经球阀时,固体颗粒会与气流分离,并在壁面上产生三处磨损集中区;随着球阀开度的减小,球阀壁面磨损量会急剧增大,且阀球内通道壁面上的磨损集中区由块状逐步转化为带状,而球阀出口处的磨损集中区则会逐渐向下移动;球阀流道直径的减小也会使得壁面磨损量增加,但磨损集中区的分布基本不变。研究结果可为进一步优化球阀流道结构以减轻其壁面磨损提供理论依据。     

Nitrogen budget and its environmental loading in an urban ecosystem with the rapid urbanisation of China

A detailed nitrogen (N) budget has been developed for an urban ecosystem based on the method of material flow analysis. How increased human activity and urbanisation influences N cycling have also been analysed. Total N input and output in the urban ecosystem of Zhengzhou City (ZUE) was calculated at 304.8?Gg was 275.3?Gg year^?1, resulting in an N accumulation of 29.5?Gg year^?1. Industry and human life activities, which respectively accounted for 43.8% and 34.2% of total N inputs and 52.6% and 29.1% of total N outputs, were the core of N flow in the urban ecosystem. Humans activities mediated more than 98% N inputs into the ZUE, 73.2% of N was released into the atmosphere and 11.7% into hydrosphere. This very large volume of released N could contribute to regional problems. High energy consumption, insufficient wastewater treatment facility practices, and low N use efficiency are the primary causes of pollution. The major challenge ahead for the urban ecosystem is how to manage high-intensity N pollutant inputs to the urban ecosystems coupled with incomplete N cycling and removal. Based on the analysis of the N budget and loading, this study also proposes an N management strategy for the ZUE.     

Interspecific association among main plant species in the unstable slope with high-frequency debris flow北大核心CSCD

To determine the characteristics of vegetation community structure and the relationship between species in the frequent watersheds of debris flow in fragile ecological environments, based on sample survey and 2 × 2 joint table techniques, we used analysis of variance test, χ2 test, Jaccard index, Pearson correlation coefficient, and Spearman's rank correlation coefficient test to study the main plant species correlations in the stable zone, instable zone, and deposit zone of the unstable slope. The analysis of variance test showed that all 45 species pairs had a significant negative correlation in the stable zone and instable zone, whereas there was no significant negative correlation in the deposit zone, which showed that the species had an independent distribution trend. The results from the different tests showed that there were 1 pair, 4 pairs, and 4 pairs from the stable zone, instable zone, and deposit zone, respectively, which had significant interspecific association under the χ2 test and the ratios of positive correlation pairs to negative correlation pairs were 0.55, 0.67, and 0.67 in the stable zone, instable zone, and deposit zone, respectively. There were 8 pairs, 5 pairs, and 5 pairs from the stable zone, instable zone, and deposit zone, respectively, which had significant interspecific association from the Pearson correlation analysis test, and the ratios of positive correlation pairs to negative correlation pairs were 0.36, 0.45, and 0.45 in the stable zone, instable zone, and deposit zone, respectively. There were 10 pairs, 6 pairs, and 9 pairs from the stable zone, instable zone, and deposit zone, respectively, which had significant interspecific association from the Spearman's rank correlation analysis test and the ratios of positive correlation pairs to negative correlation pairs were 0.5, 0.55, and 0.6 in the stable zone, instable zone, and deposit zone, respectively. The three test results showed general similarities but there were some differences. Most species pairs did not reach a significant level in the three zones and the number of negatively related species was more than the number of positively related species. In brief, this loose interspecific relationship indicates that the entire plant community was not stable, and the interspecific relationships among species are susceptible to environmental interference in the ecologically fragile areas of the debris flow basin. Therefore, rational selection and configuration of species should be applied to promote community structure development and ecological environmental improvement in vegetation restoration process areas with high-frequency debris flow. © 2018 Science Press. All rights reserved.     

大气细颗粒物对小鼠T淋巴细胞亚群的影响

大气颗粒物对肺免疫系统有潜在毒性作用,打破免疫系统平衡,大气颗粒物成分中危害首当其冲的是大气细颗粒物(PM_(2.5)),为研究大气细颗粒物引起的机体T淋巴细胞中Th1/Th2免疫失衡方向。本研究通过免疫组化实验方法检测暴露后小鼠肺组织中T淋巴细胞的表达,进一步采用流式细胞术检测大气细颗粒物气管滴注后小鼠肺脏淋巴细胞中Th1/Th2比例。暴露组小鼠肺组织免疫组化研究结果提示浸润细胞区有大量的CD4~+T细胞,中高剂量暴露组小鼠肺组织中淋巴细胞亚群Th1/Th2比例向Th1偏移。大气细颗粒物影响免疫失衡,使T淋巴细胞向Th1漂移。     

水文变异条件下潦河生态流量计算研究

气候变化和高强度人类活动改变了流域的自然水文循环过程,导致水文序列出现变异,严重影响了流域水文生态系统的稳定性.目前,基于水文变异条件下的河道内生态流量计算已成为当前变化环境下生态水文学研究的热点问题之一.以江西鄱阳湖西北部的潦河为例,采用水文时间序列变异检验方法分析径流变异性及成因,应用水文模型对水文变异后的径流进行还原,以Kolmogorov-Smirnov（K-S）、Anderson Darling（A-D）和概率点据相关系数（PPCC）3种检验法确定月径流的最优概率分布函数,进而提出水文变异条件下最适宜的潦河生态流量计算方法.结果表明:①潦河万家埠站径流在1972年发生水文变异,突变点后流域年径流增加了12%,降水量的增加和蒸发量的减少是其主要驱动要素.②采用分布式时变增益模型（DTVGM）对径流进行还原,率定期和验证期的相关系数和Nash-Sutchliffe效率系数均大于0.78,径流模拟值和实测值拟合程度较高,表明基于DTVGM进行径流还原是可行的.③基于还原后的径流,对5种概率分布函数进行拟合优度综合检验,确定逐月最优分布函数,并估算月河道生态流量.与Tennant法、最枯月平均流量法、7Q10法等方法比较,基于最优分布函数的生态流量结果更具确定性与合理性.在气候变化和人类活动引起径流变异的背景下,考虑水文变异的河道内生态需水计算方法能够更科学地体现水文变异对生态需水过程的影响,研究结果可为潦河流域水生态保护和水资源管理提供数据支撑,也可为变化环境下水资源规划和配置提供科学依据.      

弯管与盲通管冲蚀磨损对比分析研究

为了研究石油管道流向急剧改变处的冲蚀磨损问题,采用DPM模型,通过改变入口流速、颗粒质量流速、颗粒直径,对90°弯管与盲通管的流场分布及冲蚀情况进行数值模拟。结果表明:弯管与盲通管最大冲蚀速率随入口流速的增大呈指数增长,随颗粒质量流速的增大呈线性增长;在50~100 μm粒径范围内,最大冲蚀速率随粒径的增加逐渐减小,在100~300 μm粒径范围内,随粒径的增加而增大;在入口流速、颗粒质量流速、颗粒直径相同的条件下,弯管最大冲蚀速率明显高于盲通管最大冲蚀速率,盲通管的耐蚀性更强;由于流体在盲通管产生涡旋现象,增加了颗粒能量的耗散,从而减小了进入下游管线颗粒的速率,使得颗粒更易积存于盲通段形成堆积层,减小了下游管段冲蚀速率。     

隧道并行输气管道爆炸对邻管的冲击效应分析

针对隧道内输气管道泄漏发生爆炸对相邻管道的潜在威胁,基于泄漏率的求解,得到爆炸气体的扩散分布规律,确定蒸气云爆炸的TNT当量;利用LS-DYNA有限元软件建立隧道并行输气管道爆炸模型,分析在不同的泄漏尺寸、爆心距和风速下,邻管对爆炸冲击的动力学响应。基于峰值振速的经验公式的验证,表明所采用的管隧模型的可行性。结果表明:在内压和爆炸荷载的共同作用下,隧道内管道的等效应力和速度会出现多个峰值,有别于开敞空间的爆炸规律;泄漏尺寸越小、爆心距和风速越大,管道的动力响应越小;相比爆心距和风速,管道的动力响应峰值对泄漏尺寸的变化更敏感。研究成果可为管隧结构在极端情况下的事故预防和维护抢修提供理论指导。     

城市轨道交通突发大客流协同应急决策研究

为研究城市轨道交通突发大客流下的客流控制应急决策的制定与调整,对城市轨道交通客流特征进行分析,将协同矩阵与多目标规划理论运用到突发大客流下多车站协同客流控制应急决策研究中。在协同车站间路网拓扑结构的基础上,提出分别考虑各车站协同与不协同、协同车站加入或退出、协同车站评价标准变化等情况下,初始应急决策的制定与动态方案调整方法。研究结果表明:使用该方法能够表达城市轨道交通车站间协同作业的各种状态,实现突发客流冲击情况下的路网多车站动态协同应急决策的制定与动态调整,可为城市轨道交通网络化协同作业理论与方法的研究提供参考。     

公路隧道压缩空气泡沫系统灭油池火实验研究

为研究压缩空气泡沫与4.65 m2汽油池火作用过程中隧道内温度、热辐射强度、高温烟气等的变化规律,采用30 m×6 m×6 m公路隧道实验模型,考察公路隧道压缩空气泡沫系统对油池火的灭火性能。结果表明:在供给强度为5.1 L/(min·m2)、气液比14∶1条件下,公路隧道压缩空气泡沫系统对于汽油池火具有优异的控灭火能力,控火时间为21 s,灭火时间为27 s,且泡沫性能稳定,抗复燃能力强;压缩空气泡沫对于隧道内高温烟气层扰动很小,不会导致高温烟气下降到隧道下部,故不影响人员逃生疏散;在压缩空气泡沫作用下,隧道顶部及侧壁100 ℃以上高温持续时间均不超过150 s,并且可在30 s内将油池火周围的热辐射强度降至安全范围。