BP模型的改进及其在大气污染预报中的应用

针对传统BP模型存在着训练速度较慢、局部极值以及最佳网络结构无法准确确定的不足，进行了改进，应用于城市空气污染预报，建立大气污染浓度的神经网络预测模型。计算结果表明，应用改进的BP模型进行大气污染预报能够得到更好的预测结果，具有很强的实用性。     

化学品厌氧生物降解QSBR预测模型研究

收集了155种有机化学品厌氧生物降解数据,以随机抽取的109种物质作为训练集,另外46种物质作为验证集,通过结构式拆分得到各基团,分别采用多元线性回归和BP人工神经网络2种算法对有机化合物结构与生物降解性定量关系(QSBR)进行研究。结果表明,多元线性回归模型验证集正确率为78.26%,总正确率为84.52%;BP人工神经网络模型验证集正确率为82.61%,总正确率为90.32%。可见,BP人工神经网络算法相对优于多元线性回归算法。     

基于多模型优选的区域土壤重金属含量空间预测方法研究

土壤重金属含量空间预测研究对实现区域土壤资源的优化利用以及土壤环境的保护和污染防治具有重要意义。以江苏省常州市金坛区为例,基于源、汇和空间分异因子,利用BP神经网络(back-propagation network)建模方法,分别构建了源汇模型(BP-S)、空间分异模型(BP-K)和改进的多因素综合模型(BP-SK),模拟预测了区域土壤重金属Cd、Pb、Cr、Cu和Zn含量的空间分布,并对各模型预测精度进行对比分析。针对不同模型在不同区域和元素间预测精度的差异,优选出预测精度最高的模型组合,以此探求区域重金属含量空间分布最优预测结果。结果表明:(1)BP-SK模型对Cd、Cr、Cu和Zn含量预测精度均高于BP-S和BP-K模型,仅在对Pb含量的预测中BP-S、BP-K模型精度高于BP-SK模型,BP-SK模型比其他模型更能突出局部特征,包含的信息更丰富。(2)优选后最优模型预测精度较原单一模型均有不同程度的提高,对Cd、Pb、Cr、Cu和Zn含量的预测精度分别提高15.15%、20.71%、19.19%、1.75%和9.24%。(3)各模型对Cd、Pb、Cr、Cu和Zn含量的空间预测高值区均位于区域中部和东北部,低值区位于西部丘陵山区,BP-SK模型在人为影响剧烈的地区预测效果更好,而BP-K模型在自然因素影响较大的丘陵山地区的适用性更好。     

快速渗滤土地处理系统对城市生活污水中N、P去除的优化研究

为了探索城市生活污水中N、P去除，在实验室内采用混合土为介质，以均匀设计原理为指导，进行模拟人工快速渗滤系统对城市生活污水中N、P的去除。研究不同介质配比，淹水时间，湿干比3种因素组合对污水处理的最优运行模式。试验结果表明：通过选取混合土为介质及以上3个参数是可行的；通过回归统计得出最优模型。     

太阳黑子与杉木生长关系

根据多层误差板传网络结构模型和三次设计发展了一种太阳黑子人工神经网络预报方法，以杉木胸径生长的年轮指数和太阳黑子自相关因子输入变量，应用改进的人工神经网络方法建立了太阳黑子相对数年平均值的预测模型，模型的模拟回归优度为93．3％，预测精度达到95．74％，并对网络模型中变量进行灵敏度分析，分析表明，杉木胸径生长的年轮指数对太阳黑子对相对数年平均值影响较平坦，而太阳黑子自相关因子Yt-4和Yt-2对太阳子相对数年平均值影响较灵敏，3个因子对太阳黑子相对数年平均值均在一定的影响。图2表5参19     

神经网络模型森林生物量遥感估测方法的研究

森林生物量的估测是全球变化研究的基础,而遥感宏观、综合、动态、快速的特点决定了基于遥感的生物量模型为森林生物量估测的发展方向,目前的遥感生物量估测方法大多基于回归分析,需要预先假设、事后检验,仅为经验性的统计模型。神经网络的分布并行处理、非线性映射、自适应学习和容错等特性,使其具有独特的信息处理和计算能力,在机制尚不清楚的高维非线性系统体现出强大优势,可以用于遥感生物量估测。文章在野外调查的基础上,尝试应用BP网络和RBF网络技术,建立广州TM遥感影像数据与森林样方生物量实测数据之间的神经网络模型,通过训练和仿真,与生物量实测数据进行比较。结果表明,在独立样地估测中,人工神经网络估测的相对误差均小于15.18%,获得了满意的效果。而RBF网络与BP网络相比,在识别精度上、稳定性、速度上,均优于BP网络,其最大相对误差不超过10.12%,平均相对误差为4.76%。可见应用神经网络方法的“黑箱”操作虽然难以归纳出指导性规律,但可以获得很高的精度。尤其RBF网络,在训练完成后,可以应用该模型进行大区域生物量估算,对于森林的规划及管理具有深远意义。     

水安全智能识别模型的建立及应用

通过人口迁移算法优化投影寻踪模型,提出了一种新的水安全智能识别模型。与遗传算法优化的投影模型相对比,人口迁移算法的自身优势有效地避免了网络早熟现象及寻找全局最优解的困扰。从水安全的评价结果来看,用人口迁移算法优化投影寻踪是可行的,并显示出优越性。人口迁移算法为求解投影寻踪模型的非线性约束提供了新的优化方法,并为水安全评价工作提供了新的智能识别模型。     

基于遗传算法的苏云金芽孢杆菌培养基配方优化

为获取苏云金芽孢杆菌(Bacilus thuringiensis)培养基的最优配方,即玉米淀粉、黄豆饼粉、酵母粉、蛋白胨和鱼粉等的最佳配比,运用二次正交回归旋转组合设计安排试验.基于试验数据、背景知识和遗传算法的原理,进一步设计了搜索Bt培养基最优配方的算法,通过该算法搜索出该菌发酵培养基配方的最优解区间.验证性的试验结果和分析表明,基于陔遗传算法的Bt培养基配方优化的方法是有效且优于传统配方优化方法的.     

基于多元线性模型、支持向量机(SVM)模型和地统计方法的地表水溶解性总固体(TDS)估算及其精度对比——以艾比湖流域为例

地表水溶解性总固体(TDS)是地表水各组分浓度的总指标,是地表水水化学特性长期演变的最终结果,也是表征水文地球化学作用过程的重要参数,TDS的高低直接影响地表水的含盐量.本研究以艾比湖流域为研究对象,结合实测地表水TDS数据;选用准同步的Landsat OLI数据,首先,利用光谱诊断指数选取与地表水TDS相关性较高的波段,其次,利用地统计方法、多元线性回归模型和支持向量机(SVM)模型对TDS进行预测,并对其结果进行精度比较.结果表明,SVM模型为最优估测模型,拟合决定系数R2为0.97,均方误差(RMSE)为50.59;多元线性回归模型的精度与SVM模型精度较为接近,拟合决定系数R2为0.9,RMSE为66.55;地统计克里格插值法预测精度最低,拟合决定系数R2为0.87,RMSE为95.73.遥感估测SVM模型预测值在大区域能较好地反映出艾比湖流域TDS的总体特征.该模型在水质遥感领域的应用中具有良好的可行性和有效性,其预测结果也与艾比湖流域水体TDS的实际分布相吻合,因此遥感估测SVM模型在水质估测中具有一定的应用潜力.     

辽西大凌河流域土地利用变化及驱动力分析

从政策、流域综合治理、经济发展和技术进步、农民认知态度等4方面对影响大凌河流域土地利用变化的驱动力进行了分析。同时运用农户问卷调查和驱动力分析结果,选取影响耕地变化的社会经济和人口因子,运用主成分分析和多元迭代回归分析确定影响耕地变化的主要因子,并拟合出耕地变化的最优度模型。研究结果表明:在1987—2002年期间,农田和未利用荒地面积在不断减小,而林地、果园、草地在不断增加,但1995年后变化边际度大大减小;主成分分析表明影响土地利用变化主要影响因子是农业人口(A-POP)、总人口(T-POP)、农村经济收入(TIRE)、农林牧渔收入(IAFAF)和第三产业总产值(GTI);多元迭代回归分析表明耕地面积变化的最优回归模型中主变量是农业人口(A-POP)、总人口(T-POP)、农村经济收入(TIRE),这些变量能够解释95.1%的耕地变化。     

基于时间序列神经网络预测的研究

阐述了BP神经网络基本原理,并对G．P算法进行描述,提出了一种基于BP神经网络的时间序列的预测和方法,通过应用实例的分析,该方法可以得到BP网络应用于非线性时间序列预测是可行的,结果表明：神经网络方法可以成功地用于分析预测时间序列变量．参4．     

A simulation model for Japanese sardine (Sardinops melanostictus) migrations in the western North Pacific

A two-dimensional individual-based model coupled with fish bioenergetics was developed to simulate migration and growth of Japanese sardine (Sardinops melanostictus) in the western North Pacific. In the model, fish movement is controlled by feeding and spawning migrations with passive transport by simulated ocean current. Feeding migration was assumed to be governed by search for local optimal habitats, which is estimated by the spatial distribution of net growth rate of a sardine bioenergetics model. The forage density is one of the most important factors which determines the geographical distributions of Japanese sardine during their feeding migrations. Spawning migration was modeled by an artificial neural network (ANN) with an input layer composed of five neurons that receive environmental information (surface temperature, temperature change experienced, current speed, day length and distance from land). Once the weight of the ANN was determined, the fish movement was solved by combining with the feeding migration model. To obtain the weights of the ANN, three experiments were conducted in which (1) the ANN was trained with back propagation (BP) method with optimum training data, (2) genetic algorithm (GA) was used to adjust the weights and (3) the weights of the ANN were decided by the GA with BP, respectively. BP is a supervised learning technique for training ANNs. GA is a search technique used in computing to find approximate solutions, such as optimization of parameters. Condition factor of sardine in the model is used as a factor of optimization in the GA works. The methods using only BP or GA did not work to search the appropriate weights in the ANN for spawning migration. In the third method, which is a combined approach of GA with BP, the model reproduced the most realistic spawning migration of Japanese sardine. The changes in temperature and day length are important factors for the orientation cues of Japanese sardine according to the sensitivity analysis of the weights of the ANN.     

Modeling and optimization of Nile blue sulfate mineralization by heterogeneous fenton oxidation

Artificial neural network and response surface methodology have been used to develop a model for simulation and optimization of the removal of Nile blue sulfate by heterogeneous Fenton oxidation. Experimental data were used to train an artificial neural network model with linear transfer function at the output layer and a tangent sigmoid transfer function at the hidden layer. A Box–Behnken design was employed to assess the effects of input process parameters on the total organic carbon removal. First order kinetics and lumped kinetics models were used to describe the reaction; a high regression coefficient indicated that the latter fitted best. The formation of non-oxidizable compounds was shown by liquid chromatography–mass spectrometry.     

酚类化合物臭氧氧化速率的神经网络研究

酚类化合物(BP)是重要的工业原料或中间体,但工业废水含有的酚类化合物会对环境造成污染。为建立酚类化合物臭氧氧化速率的QSPR(quantitative structure-property relationship)预测模型,分析了23种酚的分子结构与臭氧氧化速率之间的相关关系,计算了这些酚的分子连接性指数和分子形状指数,优化筛选了连接性指数的1χ和2χ、分子形状指数的K1和K2共4种参数,将其作为BP神经网络的输入层变量,臭氧氧化速率作为输出层变量,采用4:2:1的网络结构,获得了令人满意的QSPR神经网络预测模型,模型总相关系数r为0.976,计算得到的臭氧氧化速率的预测值与实验值较为吻合,平均残差仅为0.05;为检验结构参数建立模型的普适性,同样方法建立对酚类化合物的辛醇-水分配系数的预测模型,模型总相关系数r达到0.993,辛醇-水分配系数的预测值与实验值吻合度较为理想,结果表明,本法建构的神经网络模型具有良好的稳健性和预测能力。     

Microsatellites and artificial neural networks: tools for the discrimination between natural and hatchery brown trout (Salmo trutta, L.) in Atlantic populations

Artificial Neural Networks (ANN) were applied to microsatellite data (highly variable genetic markers) to separate genetically differentiated forms of brown trout (Salmo trutta) in south-western France. A classic feed-forward network with one hidden layer was used. Training was performed using a back-propagation algorithm and reference samples representing the different genetic types. The hold-out and the leave-one-out procedures were used to test the validity of the network. They were chosen according to the populations and the questions analysed. The informative content of the different variables used for the distinction (the alleles of the different loci) was also evaluated using the Garson–Goh algorithm. The results of learning gave high percentages of well-classified individuals (up to 95% for the test with the hold-out analysis). This confirms that ANNs are suitable for such genetic analyses of populations. From a biological point of view, the study enabled evaluation of the genetic composition and differentiation of different river populations and of the impact of stocking.     

基于人工神经网络的城市用水需求组合预测

城市用水需求预测是涉及到诸多要素的复杂系统预测问题。为了减少简单外推法预测所带来的误差,通过在训练BP神经网络时自动调整学习步长和添加动量项修正神经单元之间的权重,既提高了神经网络的收敛速度,又抑制了神经网络限于局部极小现象的发生;然后使用改进的BP神经网络寻找多元回归预测、径向基函数(RBF)神经网络和改进BP神经网络3个单项预测的最佳组合,来综合各项独立预测所包含的信息,并以条件假设按照参考、高、低3个方案预测分析某城市的用水需求情况,说明这种基于人工神经网络的组合预测方法在预测城市用水需求量时是一个准确高效的方法。     

Neural network modeling of dissolved oxygen in the Gru?a reservoir, Serbia

The objective of this study is to develop a feedforward neural network (FNN) model to predict the dissolved oxygen in the Gru?a Reservoir, Serbia. The neural network model was developed using experimental data which are collected during a three years. The input variables of the neural network are: water pH, water temperature, chloride, total phosphate, nitrites, nitrates, ammonia, iron, manganese and electrical conductivity. Sensitivity analysis is used to determine the influence of input variables on the dependent variable. The most effective inputs are determined as pH and temperature, while nitrates, chloride and total phosphate are found to be least effective parameters. The Levenberg-Marquardt algorithm is used to train the FNN. The optimal FNN architecture was determined. The FNN architecture having 15 hidden neurons gives the best choice. Results of FNN models have been compared with the measured data on the basis of correlation coefficient (r), mean absolute error (MAE) and mean square error (MSE). Comparing the modelled values by FNN with the experimental data indicates that neural network model provides accurate results.     

Online recognition of drainage type based on UV-vis spectra and derivative neural network algorithm

• UV-vis absorption analyzer was applied in drainage type online recognition. • The UV-vis spectrum of four drainage types were collected and evaluated. • A convolutional neural network with multiple derivative inputs was established. • Effects of different network structures and input contents were compared. Optimizing sewage collection is important for water pollution control and wastewater treatment plants quality and efficiency improvement. Currently, the urban drainage pipeline network is upgrading to improve its classification and collection ability. However, there is a lack of efficient online monitoring and identification technology. UV-visible absorption spectrum probe is considered as a potential monitoring method due to its small size, reagent-free and fast detection. Because the performance parameters of probe like optic resolution, dynamic interval and signal-to-noise ratio are weak and high turbidity of sewage raises the noise level, it is necessary to extract shape features from the turbidity disturbed drainage spectrum for classification purposes. In this study, drainage network samples were online collected and tested, and four types were labeled according to sample sites and environment situation. Derivative spectrum were adopted to amplify the shape features, while convolutional neural network algorithm was established to conduct nonlinear spectrum classification. Influence of input and network structure on classification accuracy was compared. Original spectrum, first-order derivative spectrum and a combination of both were set to be three different inputs. Artificial neural network with or without convolutional layer were set be two different network structures. The results revealed a convolutional neural network combined with inputs of first and zero-order derivatives was proposed to have the best classification effect on domestic sewage, mixed rainwater, rainwater and industrial sewage. The recognition rate of industrial wastewater was 100%, and the recognition rate of domestic sewage and rainwater mixing system were over 90%.     

Unbiased sensitivity analysis and pruning techniques in neural networks for surface ozone modelling

This paper presents the use of artificial neural networks (ANNs) for surface ozone modelling. Due to the usual non-linear nature of problems in ecology, the use of ANNs has proven to be a common practice in this field. Nevertheless, few efforts have been made to acquire knowledge about the problems by analysing the useful, but often complex, input–output mapping performed by these models. In fact, researchers are not only interested in accurate methods but also in understandable models. In the present paper, we propose a methodology to extract the governing rules of trained ANN which, in turn, yields simplified models by using unbiased sensitivity and pruning techniques. Our proposal has been evaluated in thousands of trained ANNs under different conditions to establish a relationship between present contaminants (or several atmospheric variables) and surface ozone concentrations. The technique presented has demonstrated to be unbiased and stable with regard to the interpretability of the models and the good results obtained.