基于BP神经网络的矿车运行时矿井摩擦阻力的预测*

为了快速有效地确定矿车等运输设备在巷道内运行时矿井摩擦阻力的变化情况,克服模拟软件计算量和现场实测工作量大的问题,以巷道风流速度、矿车运行速度、阻塞比、矿车长度4个矿车运行时巷道摩擦阻力的影响因素作为切入点,采用动网格技术模拟得到矿车在巷道内运行时有关矿井摩擦阻力的数据,以此为样本构建基于BP神经网络的矿井摩擦阻力预测模型,运用MATLAB软件进行网络训练,并将BP神经网络预测值与FLUENT模拟值进行对比。研究结果表明:BP神经网络结构比较简单,能以较快速度收敛,预测值与模拟值最大误差在7%以内,该神经网络模型用于求解矿车等运输设备在巷道内运行时摩擦阻力的变化情况是可行的。     

Country logistics performance and disaster impact

The aim of this paper is to deepen the understanding of the relationship between country logistics performance and disaster impact. The relationship is analysed through correlation analysis and regression models for 117 countries for the years 2007 to 2012 with disaster impact variables from the International Disaster Database (EM‐DAT) and logistics performance indicators from the World Bank. The results show a significant relationship between country logistics performance and disaster impact overall and for five out of six specific logistic performance indicators. These specific indicators were further used to explore the relationship between country logistic performance and disaster impact for three specific disaster types (epidemic, flood and storm). The findings enhance the understanding of the role of logistics in a humanitarian context with empirical evidence of the importance of country logistics performance in disaster response operations.     

裂片石莼和土霉素复合压力下弧菌群落及耐药基因变化特性研究

近年来,水产养殖病源弧菌污染及其耐药性问题备受关注,本研究首次采用宏基因组测序技术分析了海水养殖沉积物在大型海藻裂片石莼（12 g·L^-1）与不同浓度（0、10、100、500 μg·L^-1）土霉素作用3个月后,弧菌群落结构及其耐药基因变化特性.结果表明,经裂片石莼诱导培养后,沉积物中弧菌属相对丰度降低了0.22%,但对弧菌多样性和弧菌耐药基因含量提升明显.裂片石莼和土霉素复合压力下,100 μg·L^-1土霉素对弧菌群落多样性及弧菌耐药基因影响最大,而500 μg·L^-1对其影响最小.沉积物弧菌属共检测出21种耐药基因,多为BL1_fox、BL2be_per、BL2_veb等β-内酰胺类耐药基因.裂片石莼能减少耐药基因tetPB富集,而进一步添加土霉素能诱导产生VatB、BcrA耐药基因.此外,弧菌耐药基因迁移与转化还受弧菌群落结构变化影响.     

低温下废铁屑对厌氧氨氧化系统的影响

考察了在低温条件下(<20℃)废铁屑及其投加方式对厌氧氨氧化反应器脱氮性能和微生物群落的影响.结果表明,当废铁屑投加量为10g/L时,直接(R2)和间接(R3)投加方式均会对厌氧氨氧化反应造成短期抑制,总氮去除率分别降低4.7%和3.4%;30d连续运行后,2组反应器总氮去除率均提升至70%左右;反应器稳定运行阶段,R2的R_s(NO₂^--N与NH₄⁺-N去除量之比)和R_p(NO₃^--N生成量与NH₄⁺-N去除量之比)为1.57和0.22, R3的R_s和R_p为1.49和0.23,比R2更接近厌氧氨氧化反应理论值.废铁屑在水中发生腐蚀,降低DO并提高pH值,且R2,R3污泥中铁含量分别为对照组的1.64倍和1.93倍,废铁屑不仅改善了厌氧氨氧菌的生境,还满足了其对铁元素的需求.高通量测序结果显示,在20~50d的运行过程中, R1,R2,R3中优势厌氧氨氧化菌属Candidatus_Kuenenia的相对丰度分别增加-1.05%,0.14%和0.96%,废铁屑的投加促进了厌氧氨氧化菌在低温下的生长,且间接投加促进效果更为显著.     

纳米零价铁耦合黏土修复污染土壤的研究进展

黏土负载型nZVI具有低成本、易制备、环境相容性好、综合性能优越的特点.本文在归纳nZVI改性策略的基础上,比较了黏土种类对nZVI形貌和性能的影响,提出了理想黏土载体的优选顺序,分析了复合材料的铁含量、比表面积、nZVI尺寸等与去除性能之间的关系,对nZVI耦合黏土在修复土壤重金属、卤代有机物、硝酸盐、新型污染物等方面的研究进行了总结,并概述了nZVI技术在土壤修复中的负面效应,最后对nZVI耦合黏土修复污染土壤的未来方向进行了展望.     

Short-term inhalation exposure evaluations of airborne antibiotic resistance genes in environments

Antibiotic resistance is a sword of Damocles that hangs over humans. In regards to airborne antibiotic resistance genes (AARGs), critical knowledge gaps still exist in the identification of hotspots and quantification of exposure levels in different environments. Here, we have studied the profiles of AARGs, mobile genetic elements (MGEs) and bacterial communities in various atmospheric environments by high throughput qPCR and 16S rRNA gene sequencing. We propose a new AARGs exposure dose calculation that uses short-term inhalation (STI). Swine farms and hospitals were high-risk areas where AARGs standardised abundance was more abundant than suburbs and urban areas. Additionally, resistance gene abundance in swine farm worker sputum was higher than that in healthy individuals in other environments. The correlation between AARGs with MGEs and bacteria was strong in suburbs but weak in livestock farms and hospitals. STI exposure analysis revealed that occupational intake of AARGs (via PM₁₀) in swine farms and hospitals were 110 and 29 times higher than in suburbs, were 1.5 × 10⁴, 5.6 × 10⁴ and 5.1 × 10² copies, i.e., 61.9%, 75.1% and 10.7% of the overall daily inhalation intake, respectively. Our study comprehensively compares environmental differences in AARGs to identify high-risk areas, and forwardly proposes the STI exposure dose of AARGs to guide risk assessment.     

Highly efficient P uptake by Fe3O4 loaded amorphous Zr-La (carbonate) oxides: Electrostatic attraction, inner-sphere complexation and oxygen vacancies acceleration effect

Bimetallic oxides composites have received an increasing attention as promising adsorbents for aqueous phosphate (P) removal in recent years. In this study, a novel magnetic composite MZLCO was prepared by hybridizing amorphous Zr-La (carbonate) oxides (ZLCO) with nano-Fe₃O₄ through a one-pot solvothermal method for efficient phosphate adsorption. Our optimum sample of MZLCO-45 exhibited a high Langmuir maximum adsorption capacity of 96.16 mg P/g and performed well even at low phosphate concentration. The phosphate adsorption kinetics by MZLCO-45 fitted well with the pseudo-second-order model, and the adsorption capacity could reach 79% of the ultimate value within the first 60 min. The phosphate adsorption process was highly pH-dependent, and MZLCO-45 performed well over a wide pH range of 2.0-8.0. Moreover, MZLCO-45 showed a strong selectivity to phosphate in the presence of competing ions (Cl⁻, NO₃⁻, SO₄²⁻, HCO₃⁻, Ca²⁺, and Mg²⁺) and a good reusability using the eluent of NaOH/NaCl mixture, then 64% adsorption capacity remained after ten recycles. The initial 2.0 mg P/L in municipal wastewater and surface water could be efficiently reduced to below 0.1mg P/L by 0.07 g/L MZLCO-45, and the phosphate removal efficiencies were 95.7% and 96.21%, respectively. Phosphate adsorption mechanisms by MZLCO-45 could be attributed to electrostatic attraction and the inner-sphere complexation via ligand exchange forming Zr/La-O-P, -OH and CO₃²⁻ groups on MZLCO-45 surface played important roles in the ligand exchange process. The existence of oxygen vacancies could accelerate the phosphate absorption rate of the MZLCO-45 composites.     

基于多元随机效应Wiener过程的某弹用弹簧可靠性评估

目的 解决某弹用弹簧可靠性评估中存在的不确定性问题。方法 对某弹用弹簧进行加速退化试验,试验过程中发现其具有多个性能退化参数,并且其退化过程具有非线性、随机性及样本差异性的特征。为了对其退化过程进行数学表征,提出一种基于多元随机效应Wiener过程的退化模型。采用基于时间尺度变换的非线性Wiener过程,描述弹簧退化过程中的非线性与随机性。基于随机效应模型,对样本的差异性进行表征,利用Copula函数描述多元性能参数之间的相互作用关系。最后推导弹簧失效寿命的概率密度函数、可靠度函数。结果 在贮存条件为25 ℃、可靠度为0.9时,基于构建的多元随机效应Wiener过程模型,当考虑性能间的相关性时,得到某弹用弹簧的贮存寿命为12 a;不考虑性能间的相关性时,得到其贮存寿命为11.5 a。结论 构建的多元随机退化模型为该弹药健康状态管理提供了相应的理论依据。     

机器学习在材料服役性能预测中的应用

针对材料服役性能预测存在误差大、计算复杂、适用性差等问题,提出了基于数据挖掘的机器学习预测方法。首先阐述了机器学习的应用流程,并总结了常用模型原理及其在材料性能预测中的应用。然后采用多种机器学习模型对RPV钢的辐照性能进行预测,并通过Stacking集成方法提高了模型的预测精度。结果表明,机器学习可用于材料服役性能预测,具有较高的预测精度和可靠性。根据材料服役数据的不同特征选择合适的学习模型,同时进行模型融合和参数优化,可有效提高模型的预测精度及运算速度。     

探索性数据分析在西藏气候变化趋势研究中的应用

为了减少小样本时间序列资料中个别异常值对气候变化趋势分析的破坏性影响,采用探索性数据分析方法之一的三组耐抗线对西藏近30年气温、云量资料进行线性趋势分析,研究了西藏地区气温和云量的变化趋势及其区域分布,并讨论了二者的相关特征。通过与传统的最小二乘法相比,表明三组耐抗线用于西藏气候变化趋势分析具有明显优势,而用最小二乘法得到的西藏气候变暖估计偏高。分析认为,西藏气候变暖主要由最低气温增高显著所致,最低气温变暖趋势在春、夏季较明显,冬季相对不明显;西藏各地增温幅度在空间分布上不一致,中、西部地区呈较强的增温趋势,东部相对较弱;西藏高原总云量和低云量总体呈减少趋势,总云量和低云量变少分别与西部和中部地区的升温之间存在显著负相关关系。