基于BP网络的水质综合评价模型及其应用

讨论了BP网络模型存在的不足及建模条件，提出了建立合理的BP网络模型的基本原则和步骤。针对水质评价问题，通过在各类水质污染指标浓度区间内生成随机分布样本的方法，组成足够多用于BP网络训练、检验和测试用的样本，建立了辽河水质综合评价的BP网络模型；给出了区分不同类别水质的模型分界值样本和模型输出分界值。     

基于IGA-BP网络的水质预测方法

针对水质预测中传统BP神经网络模型收敛速度慢,对隐层结点选取缺乏有效的手段等问题,引入了遗传算法优化BP网络的结构和隐层神经元阈值和连接权值,通过设计灵活的实数编码方案和新型交叉算子等,对实数编码遗传算法进行改进,在此基础上,提出了一种基于改进的实数编码遗传算法优化BP神经网络(IGA-BP)的水质预测新模型,并以安徽蚌埠蚌埠闸逐周水质监测的PH值数据为例,进行水质预测,通过与传统的GA-BP神经网络以及BP神经网络的水质预测模型对比,结果表明,这种预测方法训练的BP神经网络收敛速度快,样本逼近精度高且泛化能力强。     

基于人工蜂群算法与BP神经网络的水质评价模型

针对BP网络水质评价模型的不足,引入人工蜂群(ABC)算法,将求解BP神经网络各层权值、阀值的过程转化为蜜蜂寻找最佳蜜源的过程,提出了一种新的结合人工蜂群算法的BP网络水质评价方法(ABC-BP)。并以2000—2006年渭河监测断面的10组实测数据作为测试样本对其水质进行了评价,实验结果表明该方法得到的水质评价结果准确,并具有很强的稳定性和鲁棒性。     

基于集对分析的降水酸度及水质相似预测模型研究

应用集对分析法刻画历史样本之间的相似性,用多个最相似的历史样本的加权平均值作为当前样本预测值,建立了基于集对分析的降水酸度及水质相似预测模型(SFM-SPA),并运用该模型进行了降水酸度及水质预测的实例验证。结果表明,在主要影响因子选择适当、历史样本的代表性和相似性较好等限定条件下,利用模型进行环境预测是可行的,它较直观、计算简便,为环境污染预测提供了新的途径。     

基于BP神经网络的垃圾焚烧过程故障诊断

应用BP神经网络方法,建立了垃圾焚烧过程故障诊断的模型.该方法采用梯度下降法训练网络权值,利用最优停止法对网络进行了优化,避免了过拟合现象.提高了BP神经网络的训练速度和泛化能力.并以实际焚烧过程工况数据进行性能检验,检验结果表明了该BP网络具有较高的诊断精度.     

基于径向基函数网络的溢油预测模型

为了提高溢油预测的准确性,建立和优化溢油预测模型,提出了基于径向基函数网络模型的溢油预测方法,实现溢油预测功能.径向基函数网络模型解决了模拟预测过程中样本库巨大、函数模型收敛速度慢的问题.通过选择有效的输入参数和样本数据,建立局部逼近网络;通过径向基函数训练样本数据,利用输出值与实际值之间的误差作为约束条件调整权重因子、径向基中心和宽度,加快函数模型的收敛速度.该模型模拟了溢油的漂移、扩散过程,达到预测的目的.利用该模型,建立了溢油预测模块,并针对一次溢油事故进行预测模拟,验证了该模型的可行性,能够为应急决策提供一定的支持.     

Carrousel氧化沟系统出水COD预报的神经网络模型

以河南漯河市污水净化中心的氧化沟系统为考察对象,针对该系统进水水质复杂,控制滞后的难点,引入人工神经网络的理论和方法,对其进行模拟分析,建立了基于BP网络的氧化沟系统出水COD预报模型。模型性能检验和灵敏度检验表明,建成的模型准确度高,适应性强,可直接用于该系统出水COD预报。这为氧化沟工艺在线控制提供了一条简便的途径。     

基于灰色模型和模糊神经网络的综合水质预测模型研究

水质状态变化趋势预测研究对水资源管理和维护具有重要的现实意义。提出了一种将灰色模型和模糊神经网络相结合的水质预测模型。首先基于改进的灰色模型预测出水体中各理化因子在未来一段时间内的指标变化,然后采用T-S模糊神经网络对各单因子的预测值进行数据融合,构建水质变化综合趋势预测模型,预测出下一时间段的水质整体状态指标。实验表明,这种方式用来预测湖泊水质变化趋势具有可行性;与BP网络模型相比,基于T-S模糊神经网络系统的模型具有预测精度高、模型系统稳定等优越性。     

变权组合模型在景观水体水质模拟中的应用

针对景观水体的水质模拟与预测问题,在BP神经网络和支持向量机模型的基础上,建立了权重随输入量变化的变权组合模型。该模型既能充分利用各个单一模型的优点,又能避免固定权重分配的弊端。经实例验证,与单一的BP神经网络和支持向量机模型相比,变权组合模型拟合精度更高,预测结果更为准确。     

利用复介电常数和BPAT神经网络进行环境监测的可行性研究

土壤中的污染物成分复杂，其含量与复介电常数之间具有很强的非线性关系。以土壤样品复介电常数的实部、虚部分别作为输入，以其含水率、体密度和所含6种已知离子的浓度分别作为输出，建立BP人工神经网络。把吉泰兰地区的土壤样品数据分为训练样本集和检验样本集，网络训练后，其学习效果显示模型的性能很好，检验样本的预测结果也与实测值较好吻合，说明利用复介电常数和BP人工神经网络进行环境监测是一种好的方法。     

Anatomy of an urban waterbody: a case study of Boston's Muddy River

The objective of this study was to characterize and understand the water quality of Boston's Muddy River prior to restoration, to help guide those activities and evaluate their success. We use a combination of monitoring, data analysis and mathematical modeling. The seasonal pattern of temperature, pollutant signatures (identified using a principal component analysis), correlations with precipitation and spatial patterns all point to a significant wastewater input at one of the outfalls and suggest significant receiving water impact. However, a quantitative analysis using a mathematical model (QUAL2K) suggests this source is not significant. Rather, internal loading from algae, sediment bed and waterfowl dominate the spatial pattern of water quality. These results suggest significant improvement can be expected from planned sediment dredging. The paper provides a case study of water quality assessment in the context of urban river restoration, and it illustrates the utility of combining monitoring and data analysis with modeling.     

瘦西湖风光电能驱动的曝气生物接触氧化水净化系统结构与功能

在瘦西湖选择中试河段,利用太阳能风能绿色能源,研发了风光电能驱动的曝气生物接触氧化水净化系统。系统由风光互补发电子系统、曝气生物接触氧化子系统、植物生态子系统组成。风光电能驱动的曝气生物接触氧化水净化系统运行正常。按照景观娱乐用水水质标准进行水质监测,采用模糊综合评价与贴近度计算方法,进行了系统水净化功能评价。结果表明,在试验段原水水质低于景观水C类,经该系统水净化后,可以达到景观水B类。系统的开发与实验,探索了利用绿色能源河湖水净化的新途径。     

程序化的未确知测度模型用于水环境质量评价简

未确知测度模型用于水环境质量评价具有合理性。为使模型更加完善,提出将未确知测度模型建立过程标准化,使其具有通用性。并针对水质评价指标众多致评价过程复杂、计算量大的情况,利用MATLAB软件将标准化后的未确知测度模型设计为既定的程序,依靠计算机模拟来实现水质评价。通过对浮桥河水库水质的评价结果及与其他多种方法的对比,表现出标准化模型的客观性、合理性,MATLAB程序的实用性、快捷性。程序化的未确知测度模型标准化后可进行推广。     

程序化的未确知测度模型用于水环境质量评价

未确知测度模型用于水环境质量评价具有合理性。为使模型更加完善,提出将未确知测度模型建立过程标准化,使其具有通用性。并针对水质评价指标众多致评价过程复杂、计算量大的情况,利用MATLAB软件将标准化后的未确知测度模型设计为既定的程序,依靠计算机模拟来实现水质评价。通过对浮桥河水库水质的评价结果及与其他多种方法的对比,表现出标准化模型的客观性、合理性,MATLAB程序的实用性、快捷性。程序化的未确知测度模型标准化后可进行推广。     

改进的QGA-BP模型在弥苴河总氮量预测中的应用

水质预测对水环境规划、评价和管理十分重要。构建一种改进的量子遗传算法（QGA）优化BP神经网络的模型,即在量子遗传算法中引入了旋转角的动态改进策略和遗传算法的交叉变异操作,并以改进的QGA作为进化操作准则优化BP模型的权值和阈值。以弥苴河复杂水环境水质预测为实例,选取一组历史观测数据作为训练样本,对其进行分析。将结果与BP模型、QGA-BP模型仿真结果进行了对比,改进后的QGA-BP模型在进化代数、收敛速度和预测结果的准确率有较大提高。对弥苴河水质的预测结果表明,将改进QGA-BP模型用于水质预测是可行、有效的预测方法。     

嫩江水污染物排放总量分配方法研究

本文针对嫩江沿岸采用氧化塘处理污水的特点，为充分合理利用水环境容量，以流域总量控制的原则为指导，建立了水污染物总量分配模型，预测不同流量和不同排污情况下的水质，确定不同流量下给定水质目标的临界距离和允许排污量。     

Preliminary results of modeled ozone uptake for Fagus sylvatica L. trees at selected EU/UN-ECE intensive monitoring plots

The objective of this study was to establish whether EU and UN-ECE/ICP-Forests monitoring data (i) provide the variables necessary to apply the flux-based modeling methods and (ii) meet the quality criteria necessary to apply the flux-based critical level concept. Application of this model has been possible using environmental data collected from the EU and UN-ECE/ICP-Forests monitoring network in Switzerland and Italy for 2000-2002. The test for data completeness and plausibility resulted in 6 out of a possible total of 20 Fagus sylvatica L. plots being identified as suitable from Switzerland, Italy, Spain, and France. The results show that the collected data allow the identification of different spatial and temporal areas and periods as having higher risk to ozone than those identified using the AOT40 approach. However, it was also apparent that the quality and completeness of the available data may severely limit a complete risk assessment across Europe.     

Use of historical assessment for evaluation of process-based model projections of future environmental change: Lake acidification in the Adirondack mountains, New York, USA

Because of the considerable uncertainties associated with modeling complex ecosystem processes, it is essential that every effort be made to test model performance prior to relying on model projections for assessment of future surface water chemical response to environmental perturbation. Unfortunately, long-term chemical data with which to validate model performance are seldom available. The authors present here an evaluation of historical acidification of lake waters in the northeastern United States, and compare historical changes in a set of lakes to hindcasts from the same watershed model (MAGIC) used to estimate future changes in response to acidic deposition. The historical analyses and comparisons with MAGIC model hindcasts and forecasts of acid-base response demonstrate that the acidic and low-ANC lakes in this region are responsive to strong acid inputs. However, the model estimates suggest lakewater chemistry is more responsive to atmospheric inputs of sulfur than do the estimates based on paleolimnological historical analyses. A 'weight-of-evidence approach' that incorporates all available sources of information regarding acid-base response provides a more reasonable estimate of future change than an approach based on model projections alone. The results of these analyses have important implications for predicting future surface water chemical change in response to acidic deposition, establishing critical loads of atmospheric pollutants, and other environmental assessment activities where natural variation often exceeds the trends under investigation (high noise-to-signal ratio). Under these conditions, it is particularly important to evaluate future model projections in light of historical trends data.     

Predictions of U.K. regulated power station contributions to regional air pollution and deposition: a model comparison exercise

Contributions of the emissions from a U.K. regulated fossil-fuel power station to regional air pollution and deposition are estimated using four air quality modeling systems for the year 2003. The modeling systems vary in complexity and emphasis in the way they treat atmospheric and chemical processes, and include the Community Multiscale Air Quality (CMAQ) modeling system in its versions 4.6 and 4.7, a nested modeling system that combines long- and short-range impacts (referred to as TRACK-ADMS [Trajectory Model with Atmospheric Chemical Kinetics-Atmospheric Dispersion Modelling System]), and the Fine Resolution Atmospheric Multi-pollutant Exchange (FRAME) model. An evaluation of the baseline calculations against U.K. monitoring network data is performed. The CMAQ modeling system version 4.6 data set is selected as the reference data set for the model footprint comparison. The annual mean air concentration and total deposition footprints are summarized for each modeling system. The footprints of the power station emissions can account for a significant fraction of the local impacts for some species (e.g., more than 50% for SO2 air concentration and non-sea-salt sulfur deposition close to the source) for 2003. The spatial correlation and the coefficient of variation of the root mean square error (CVRMSE) are calculated between each model footprint and that calculated by the CMAQ modeling system version 4.6. The correlation coefficient quantifies model agreement in terms of spatial patterns, and the CVRMSE measures the magnitude of the difference between model footprints. Possible reasons for the differences between model results are discussed. Finally, implications and recommendations for the regulatory assessment of the impact of major industrial sources using regional air quality modeling systems are discussed in the light of results from this case study.     

基于氨氮实测结果的水质自动监测系统数据质量分析

为提高水质自动监测数据质量,在地表水水质自动监测系统实际监测运行环境中,基于统计分析方法,探讨水质自动监测日常运行维护质控手段对运行维护质量监督考核结果的影响,探索提高水质自动监测质控考核结果合格率的运行维护质控技术要求。结果表明：零点漂移检查、量程/跨度漂移检查对水质自动监测质控考核结果虽无显著影响,但却是保障水质自动监测数据质量的基础;水质自动监测仪标准物质核查结果相对误差的合格判定标准由±10%加严至±6%时,水质自动监测仪标准物质核查、实际水体样品比对测试、实际水体样品加标回收率测试及盲样考核结果的合格率均显著提升;标准物质核查合格率由74.6%~77.9%提升为98.4%~100%,实际水体样品比对测试的合格率由62.5%提升为75.9%,实际水体样品加标回收率测试的合格率由75.0%提升为87.5%~100%,盲样考核合格率由75.0%提升为100%;标准物质核查及实际水体样品比对测试相对误差均不服从正态分布,且数据分布为正偏态分布;实验所用水质自动监测仪器可能存在负偏离的系统误差,手工比对实验环节对实际水体样品比对测试结果也可能存在一定干扰。综合上述结果,在进行水质自动监测仪器设计和选型时,应消除可能的系统偏差;在开展实际水体样品比对测试过程中,应加强对比对实验相关环节的质量监督和检查;在水质自动监测系统运行维护时,应采用更严格的标准物质核查结果相对误差合格判定标准,以提高和保障自动监测数据质量。相关研究成果可为水质自动监测系统运行维护采取的质量控制措施和评价标准的制定提供技术支持。