基于数据分解的AQI的CEEMD-Elman神经网络预测研究

针对Elman神经网络在预测空气质量指数（AQI）时易受到数据非平稳性的影响导致预测趋势良好但准确度较低的问题,提出以互补集合经验模态分解（Complementary Ensemble Empirical Mode Decomposition,CEEMD）为基础的CEEMD-Elman模型.应用CEEMD对AQI序列分解成不同时间尺度上的本征模态函数分量和剩余分量,进而首次将对非平稳的AQI序列的预测研究转化为对多个平稳的本征模态函数分量的研究.分别与Elman单一模型、EMD-Elman模型、BP单一模型及CEEMD-BP模型进行实验对比.结果表明：应用该方法建立的模型的均方误差、平均绝对误差和平均绝对百分比误差分别为4.80、0.71、1.84%,均小于其他模型结果;对应空气质量等级预报正确天数的频率为94.12%.该模型能有效的降低非平稳性对实验预测结果的影响,实现对空气质量等级的准确预报;该研究为进一步预测AQI的走向提供了有效依据,也为政府决策和管理部门制定空气污染控制提供了更充分的参考.     

An assessment of how distance and diesel oxidation catalyst will impact thermal decomposition behaviors of particles

Decomposition mass loss and pyrolysis products analyses of particles sampled at various locations along the tailpipe of a Euro-IV diesel engine were performed using a thermogravimetry in conjunction with Fourier transformation infrared spectrometry-mass spectrum. Diesel particles were collected at the same location with and without diesel oxidation catalyst (DOC) mounted on the test engine separately. The three poles in thermal gravity-differential thermal gravity images suggested that the decomposition process of diesel particles could be divided into three stages which correspond to the decompositions of lower boiling substances, higher boiling substances and soot respectively. It is noticed that no matter whether DOC was mounted or not, the further the particles were sampled away from the engine block, the lower the peak temperatures and the heavier the mass losses within the first two stages, which indicated that the soluble organic fraction in the particle samples increased and therefore lowering the activation energy of thermal decomposition. Hydroxyl, ammonia, C_xH_y fragments, benzene, toluene, and phenol were found to be the primary products of particle decomposition, which didn't change with the location of particle sample point. The employment of DOC increased the activation energy for particle oxidation and resulted in a higher peak temperature and lower mass loss within the first-stage. Moreover, the CO stretching bands of aldehyde and ketone at 1771 cm^?1 was only detected without a DOC, while the NO₂ peak at 1634 cm^?1 was solely noticed with the presence of DOC. Compared to the first-stage pyrolysis products, more polycyclic aromatic hydrocarbons and less C_xH_y fragments were seen in the second-stage.     

藻类对有机污染物生物吸附的研究进展

随着工业的迅猛发展,越来越多的有毒有机污染物被排放到环境中,进而引起一系列环境问题。大量研究表明,生物吸附已经成为去除有毒有机污染物的重要方法,而藻类作为一种理想的生物吸附剂,具有良好的吸附和降解有机污染物的能力。概述了藻类对染料、酚类、多环芳烃三种具有代表性的有毒有机污染物的生物吸附。     

Do Nutrients Limit Algal Periphyton in Small Blackwater Coastal Plain Streams?1

Abstract: We examine the potential for nutrient limitation of algal periphyton biomass in blackwater streams draining the Georgia coastal plain. Previous studies have investigated nutrient limitation of planktonic algae in large blackwater rivers, but virtually no scientific information exists regarding how algal periphyton respond to nutrients under different light conditions in smaller, low‐flow streams. We used a modification of the Matlock periphytometer (nutrient‐diffusing substrata) to determine if algal growth was nutrient limited and/or light limited at nine sites spanning a range of human impacts from relatively undisturbed forested basins to highly disturbed agricultural sites. We employed four treatments in both shaded and sunny conditions at each site: (1) control, (2) N (NO₃‐N), (3) P (PO₄‐P), and (4) N + P (NO₃‐N + PO₄‐P). Chlorophyll a response was measured on 10 replicate substrates per treatment, after 15 days of in situ exposure. Chlorophyll a values did not approach what have been defined as nuisance levels (i.e., 100‐200 mg/m²), even in response to nutrient enrichment in sunny conditions. For Georgia coastal plain streams, algal periphyton growth appears to be primarily light limited and can be secondarily nutrient limited (most commonly by P or N + P combined) in light gaps and/or open areas receiving sunlight.     

基于ICEEMDAN和MC-CNN的矿山声发射信号识别分类方法

为精准识别地下矿山声发射事件,采用基于改进的完全集合经验模态分解模型(ICEEMDAN)和多通道卷积神经网络(MC-CNN)模型对声发射信号进行处理后得到分量图,根据各通道输入分量峭度值赋予不同权重,并利用卷积神经网络对输入数据进行训练,最终采用五折交叉实验方法验证该分类识别方法的可行性及有效性.结果表明:基于ICEE...     

超声波处理石化厂剩余活性污泥

采用超声波处理石化污水处理厂的剩余活性污泥。介绍了超声波对污泥的作用机理，初步研究了超声波处理污泥的影响因素。大功率超声波可以降解生物污泥，释放出其中的有机物。小功率超声波能够改善污泥的膨胀特性、提高污泥沉淀特性和脱水能力、降低污泥的含水率，达到减量的目的。经超声波处理并简单过滤的污泥滤液，COD达到4200mg／L左右，污泥含水率由94％降至85％左右。     

整合消费与增加值视角的中国行业碳排放变动分解研究

在研究影响我国CO 2排放因素领域,基于投入产出技术的分解模型已成为主要的分析工具,现有研究多分别基于消费视角或收益视角展开分析。为全面评估各行业收益与消费对其上游、下游行业碳排放的综合影响,整合基于收益与基于消费两个视角,运用2012年与2015年我国投入产出表,构建两层嵌套式结构分解分析模型(SDA),比较分析消费规模效应、收益规模效应、行业流入、流出增加值变动效应、增加值结构变动效应等14个影响各行业碳排放变动的关键因素,并借助对消费者原则碳排放估算公式的重构,更准确地实现从增加值视角对各行业消费者原则碳排放变动的关键影响因素分析。研究发现:①研究期内,我国在总产出增长29.14%的同时碳排放量上升1.46%,各行业碳排放强度下降是主要的减排因素,其中建筑业减排贡献最大。②增排效应方面,影响从大到小依次为消费规模、收益规模、完全投入结构与完全消费结构四项效应,且前两个规模效应的影响是后两个结构效应的2倍以上,尤其建筑业消费规模效应、煤炭采选产品业收益规模效应增排较大。③消费规模扩大导致增排的原因并非行业本身生产规模扩大,而主要在于建筑业、服务业等行业规模扩张时吸收其他行业流入的增加值量增多。④收益规模扩大导致的增排效应方面,细分来说从大到小依次为劳动者报酬、生产税净额、固定资产折旧、营业盈余四项效应,且行业差异显著,如煤炭采选产品业的劳动者报酬效应,石油、炼焦产品和核燃料加工品业的生产税净额效应以及电力、热力生产和供应业的营业盈余效应增排贡献较大,而煤炭采选产品业的营业盈余效应、批发、零售业和住宿、餐饮业的生产税净额效应则减排贡献较大。     

Internal eutrophication: How it works and what to do about it—a review

In the 1980s and 1990s, it became increasingly clear that changes in external nutrient loads alone could not entirely explain the severe eutrophication of surface waters in the Netherlands. Nowadays, 'internal eutrophication' has become a widely accepted term in Dutch water management practice to describe the eutrophication of an ecosystem without additional external input of nutrients (N, P, K). This review surveys the principal mechanisms involved in this process. It also discusses possible remedies to combat internal eutrophication.     

Changes in the Biochemical Composition of Tetraselmis Suecia and Isochrysis Galbana During Growth and Decay

The variations of the biochemical composition of Tetraselmis suecica and Isochrysis galbana during growth and decay were determined. the content of chlorophyll a (Chl-a) of the cultures, as expected, slowly degraded into phaeopigments during decay, confirming that chlorophyll measurements do not always provide an accurate estimate of phytoplanktonic biomass and, consequently, may fail if used to measure the food availability of particulate matter for consumers. Measurements of total amounts of proteins, carbohydrates and lipids, related to the nutritional value of particles in terms of caloric content, are shown to provide information on the readily available food for consumers, particularly during the blooms. the protein/carbohydrate, C/N and POC/Chl-a ratios were used to evaluate the differences between these two species during the growth and the decomposition processes. A comparison between experimental and field conditions was undertaken to implement our understanding of the growth and degradation processes of particulate organic matter of phytoplanktonic origin in the sea and its role on natural systems, during and after phytoplankton blooms.     

滇池水体除藻材料的除藻作用试验研究

利用昆明地区的粘土矿制作新型轻质除藻材料,试验研究该材料对滇池湖水中藻类的除藻及净化作用.实验室小试(20 L)结果表明,该除藻材料放入滇池水样实验70 d后,水质明显好转,藻类去除率可达95.8%,CODMn、TN及TP的去除率分别为34.4%、81.5%、28.2%.在滇池湖边300 m2水面围栏扩大试验结果显示,除藻材料放入试验区水体10 d后,水体中藻类含量明显减少,水体透光率从55%增加到80%,水质状况也有很大改善.研究结果表明,该材料具有良好的湖泊水体除藻及净化作用,有必要对其开展进一步的深入研究.