基于水质水量的城镇污水处理厂工艺运行稳定性研究

采用四川省2座位于不同排水服务区域的城镇污水处理厂的样本数据,统计分析了日处理水量,COD、TN、TP浓度及负荷和C/N、C/P等统计量的累积频率分布规律,并分析了各统计量的分布特征对工艺运行稳定性的影响。结果表明,污水处理厂A进水污染物负荷的累积频率分布范围较大,其采用的周期循环活性污泥(CASS)工艺基本能适应进水水质的随机变化,但因存在碳源随机性不足情况,可能导致其脱氮稳定性降低;污水处理厂B进水污染物负荷的累积频率分布范围较小,虽然其采用的厌氧—缺氧—好氧(A2/O)工艺的调控能力较低,但也基本能适应其服务区域内进水水质的随机变化;2座污水处理厂的日处理水量累积频率分布与污染物浓度及负荷的累积频率分布之间均存在显著的差异,日处理水量的累积频率分布比较集中,而污染物浓度及负荷的累积频率分布的离散范围较大,这也潜在威胁到工艺运行的稳定性。总体来看,2座污水处理厂的处理工艺选择恰当,均能较好适应其服务区域的进水水质和水量的要求。     

电化学氧化法处理垃圾渗滤液纳滤浓缩液

实验利用电化学氧化法处理垃圾渗滤液纳滤浓缩液,以提高废水的可生化性。研究考察了水力停留时间、进水流量、循环流量、电流强度和原水氯离子浓度对有机物去除的影响。研究结果表明,电化学氧化法的最佳运行条件如下:水力停留时间为3 h,进水流量为1 m3/h,循环流量为15 m3/h,电流强度为420 A。在上述条件下,原水COD浓度从3 100mg/L降到1 311.3 mg/L,去除率达到57.7%,BOD/COD值由0.03提升至0.31。氯离子对电解有促进作用,但原水氯离子浓度超过5 000 mg/L,不需要外加工业盐。     

不同诱导条件对结晶产物形貌特征的影响

在流化床反应器中,以含铜废水为处理对象,重点考察了诱导条件改变对结晶系统运行效率及结晶产物形貌特征的影响。结果表明,进水200 mg/L,进药比([Cu2+]/[CO2-3])1∶1.2,p H为10.2的条件下,铜离子去除率可接近100%,诱导晶种表面平滑,诱导颗粒生长成致密杆状晶体结构,以碳酸盐为沉淀剂的诱导结晶系统依赖于体系的p H,高p H体系下发生的共结晶现象是影响系统运行的主要原因;随着进药摩尔比的增加,结晶产物由短杆状向球状转化,晶体机械强度显著降低大量破碎,影响出水水质;进药比1∶1.2,p H为11的条件下,含铜废水浓度为500 mg/L时,系统去除率可达95%以上,微晶产率低于5%。     

MnO_2/Al_2O_3吸附草甘膦及微波紫外耦合降解再生工艺

以掺混煅烧法制备Mn O2/Al2O3,并对草甘膦进行吸附,将吸附态草甘膦置于微波紫外耦合系统中进行降解和再生。研究并给出了Mn O2/Al2O3的最佳制备条件、最佳吸附条件和微波紫外耦合降解再生的最佳工艺参数。Mn O2/Al2O3最佳制备条件为Mn O2质量分数15%,煅烧温度500℃,煅烧时间2 h。在常温下,Mn O2/Al2O3对草甘膦的最大吸附量为75 mg/g。微波紫外耦合系统最佳工艺参数为功率500 W,时间25 min,空气量0.06 m3/h。在此操作条件下,Mn O2/Al2O3的再生率达到85%,并且可以多次再生利用,草甘膦最终降解产物为二氧化碳、氮氧化物、磷酸和水,矿化率为65%。     

通风口高度对太阳能烟囱建筑自然通风和自然排烟性能的影响研究

安装太阳能烟囱后的建筑具有效果优良、节能环保的自然通风系统。室内通风口高度是太阳能烟囱建筑性能的关键参数之一。提出通过改变太阳能烟囱建筑室内通风口高度位置,使太阳能烟囱既可用于建筑的日常自然通风,也能实现室内火灾场景下的自然排烟。利用缩比例实验台研究了不同室内通风口高度对太阳能烟囱建筑自然通风与自然排烟效果的影响,发现自然通风和自然排烟效果随通风口高度增加呈现先增大后减小的趋势,且最佳室内通风口高度位于窗户上檐位置附近。研究结果对太阳能烟囱建筑的建筑结构关键参数设计具有指导意义。     

An automated (novel) algorithm for estimating green vegetation cover fraction from digital image: UIP-MGMEP

Green vegetation cover fraction (VCF) is an important indicator of vegetation status in ecology and agronomy. Digital image analysis (DIA) has been widely accepted as a new VCF measurement technique. In this study, we present a novel fully automatic threshold segmentation algorithm for VCF measurements, which is named as upper inflection point plus mean gradient magnitude of edge pixels (UIP-MGMEP). The algorithm performs VCF estimation upon the vegetation index Excess Green (EXG). UIP-MGMEP optimizes the EXG threshold by searching the upper inflection point (UIP) of the M-Et curve (mean gradient magnitude of edge pixels (MGMEP) vs. EXG threshold), based on the assumption that EXG variance of the boundary pixels between vegetation and background is larger than the variance of the background. Five typical sample images are used to illustrate how ground complexity reduces the distinctness of the UIP. Three controlled experiments are illustrated to test the robustness of UIP-MGMEP to resolution, exposure, and ground complexity. The results show that UIP-MGMEP is a promising algorithm for automatic VCF estimation upon digital images. Compared to broad-leaved grass, narrow-leaved grass is more sensitive to resolution and exposure. To reduce ground complexity, smaller footprint size while more images to cover the same area may be better than one image with large footprint size. UIP-MGMEP is fully automatic, making it promising for batch processing of VCF measurements that is very difficult in any wide-range field survey in the past. UIP-MGMEP algorithm can only extract green vegetation and is not suitable for non-green (even grayish-green) vegetation, due to the limits of vegetation index EXG. In addition, UIP-MGMEP is not recommended for images with VCF less than 0.5% or greater than 99.5%.     

Deep learning architecture for air quality predictions

With the rapid development of urbanization and industrialization, many developing countries are suffering from heavy air pollution. Governments and citizens have expressed increasing concern regarding air pollution because it affects human health and sustainable development worldwide. Current air quality prediction methods mainly use shallow models; however, these methods produce unsatisfactory results, which inspired us to investigate methods of predicting air quality based on deep architecture models. In this paper, a novel spatiotemporal deep learning (STDL)-based air quality prediction method that inherently considers spatial and temporal correlations is proposed. A stacked autoencoder (SAE) model is used to extract inherent air quality features, and it is trained in a greedy layer-wise manner. Compared with traditional time series prediction models, our model can predict the air quality of all stations simultaneously and shows the temporal stability in all seasons. Moreover, a comparison with the spatiotemporal artificial neural network (STANN), auto regression moving average (ARMA), and support vector regression (SVR) models demonstrates that the proposed method of performing air quality predictions has a superior performance.     

Distribution of antibiotic-resistant bacteria in chicken manure and manure-fertilized vegetables

Veterinary manure is an important pollution reservoir of antibiotics and antibiotic-resistant bacteria (ARB). However, little is known of the distribution of ARB in plant endophytic bacteria and the number/types of ARB in chicken manure. In this study, 454-pyrosequencing was used to investigate the distribution and composition of ARBs in chicken manure and fertilized vegetables. The prevalence of ARB in the samples of the chicken manure compost recovered from farms on which amoxicillin, kanamycin, gentamicin, and cephalexin were used was 20.91–65.9 % for ARBs and 8.24–20.63 % simultaneously resistant to two or more antibiotics (multiple antibiotic resistant bacteria (MARB)). Antibiotic-resistant endophytic bacteria were widely detected in celery, pakchoi, and cucumber with the highest rate of resistance to cephalexin. The pyrosequencing indicated that the chicken manure dominantly harbored Firmicutes, Bacteroidetes, Synergistetes, and Proteobacteria and that Bacteroidetes was significantly enhanced in farms utilizing antibiotics. In the total cultivable colonies, 62.58–89.43 % ARBs and 95.29 % MARB were clustered in Bacteroidetes with the dominant species (Myroides ordoratimimus and Spningobacterium spp., respectively) related to human clinical opportunistic pathogens.     

柴油机微粒中可溶性有机成分的真空干燥条件

柴油机碳烟颗粒(PM)中可溶性有机组分(SOF)含量是其重要的特性参数,而真空干燥法可方便、高效、可靠地分析颗粒SOF含量,对其干燥条件的探索具有重要的研究意义和应用价值。首先采用模拟的柴油机颗粒,分析不同的干燥条件对模拟SOF(柴油、润滑油)挥发率的影响规律,得出模拟的SOF分析的最佳干燥条件;在此基础上,探索了真实柴油机颗粒SOF分析的最佳干燥条件为:干燥温度220℃、干燥时间3 h、真空度98 kPa,得知SOF组分干燥的难易程度介于柴油和润滑油之间;同时,根据此干燥条件,分析了排气管末端采集的真实碳烟颗粒的SOF含量特性。     

煅烧的水滑石同时去除水体中砷和氟

为探讨水滑石类材料对水体中氟、砷离子的同时去除效果,采用共沉淀法合成(Mg∶Al=2∶1)纳米类水滑石(LDHs),用傅立叶转换红外光谱、电子扫描透射电镜、X射线晶体衍射等手段对合成的材料进行了表征,并研究纳米材料在不同初始浓度、pH、吸附时间、阴离子干扰条件下其同时除砷氟性能。结果表明,煅烧后的水滑石(LDOs)对砷最大吸附量为51.02 mg/g,对氟最大吸附量为36.63 mg/g。吸附动力学实验表明,煅烧水滑石对砷的吸附在前6 h内基本完成,对氟的吸附在前10 h内基本完成。砷氟共存溶液保持pH=4~10及pH=6~8时,水滑石分别对砷、氟保持良好的吸附效率。对比不同阴离子对水滑石共除砷氟效率的影响,水滑石除砷速率受到阴离子影响力大小为:HPO2-4CO2-3NO-3Cl-SO2-4;水滑石除氟速率受到阴离子影响力大小为:CO2-3HPO2-4SO2-4Cl-NO-3。材料再生循环利用4次后,对砷和氟的吸附效率均能达到90%以上。实验结果表明,所合成的水滑石是一种优秀的能共除砷氟的吸附剂。