基于小波神经网络的天然水体溶解氧预测研究

建立基于小波神经网络的预测模型,以不同时间滞差和影响因子组合作为输入变量,对海河流域四个监测断面的溶解氧浓度进行短期预测.结果表明,基于溶解氧历史数据的小波神经网络预测模型精度更高,可用于天然水体的水质预测,为水质管理提供更客观的参考和依据.     

混沌在水质不确定性预测中的应用

在Dobbins-Camp模型的基础上,利用混沌理论分析了水质预测中的误差,得到关于误差时间序列的相空间,在此基础上,将混沌神经网络嵌入Dobbins-Camp模型,建立一个具有混沌特性和学习功能的水质不确定性模型,仿真结果表明,方法在预测水质不确定性上是有效的.     

基于混沌最小二乘支持向量机的时间序列预测研究

为了对这种具有非线性特性的时间序列进行预测,提出一种基于混沌最小二乘支持向量机.算法将时间序列在相空间重构得到嵌入维数和时间延滞作为数据样本的选择依据,结合最小二乘法原理和支持向量机构建了基于混沌最小二乘支持向量机的预测模型.利用此预测模型对栾城站土壤含水量时间序列进行了预测.结果表明,经过相空间重构优化了数据样本的选取,通过模型的评价指标,混沌最小二乘支持向量机的预测模型能精确地预测具有非线性特性的时间序列,具有很好的理论和应用价值.     

基于Volterra自适应方法的水文混沌时间序列预测

Volterra泛函级数能够描述具有响应和记忆功能的非线性行为,一般用于非线性系统因果关系点对的预测,把Volterra自适应方法应用于水文混沌时间序列的预测研究是一个有意义的工作。论文针对水文系统的复杂性,基于混沌动力系统相空间重构技术,构建了水文混沌时间序列Volterra自适应预测方法,并采用NLMS算法调整滤波器参数,并就模型进行仿真计算,讨论了模型参数对预测精度的影响。直门达水文站月蒸发量混沌时间序列预测实验表明,水文混沌时间序列Volterra自适应预测方法,具有较好的预测精度和效果,拓展了水文预测报方法的研究途径。     

基于多元混沌时间序列的油田产量预测模型

油田产量的预测一直是石油工作者研究的重要课题.针对油田产油量、产水量、地层压力和时间之间有着混沌的特征,利用多变量混沌时间序列等方法研究了油田产量的混沌建模和预测问题.用C-C算法确定每一个变量的嵌入维数和延迟时间,重构多元混沌时间序列的相空间;使用基于奇异值分解的主成分分析消除重构相空间的冗余变量和噪声干扰,建立了有较好泛化性能的多元混沌时间序列油田产量预测模型;最后将混沌时间序列预测和Elman神经网络进行耦合,创建了基于主成分分析前馈网络的多元混沌时间序列油田产量预测方法.应研究表明,提出的多变量混沌时间序列预测方法的预测精确度优于单变量预测,它可用于解决具有多变量混沌时间序列的预测问题.     

基于支持向量回归机的水文混沌时间序列预测

支持向量机在系统辨识和分类研究方面比较成熟,目前尚没有提出有效的支持向量回归理论来解决非线性、时变、干扰的复杂问题.支持向量回归机主要用于因果关系点对的回归预测,把支持向量回归机应用于水文混沌时间序列的预测研究是一个有意义的工作.在支持向量机一般理论基础上,提出了水文混沌时间序列支持向量回归机模型,并就模型进行仿真计算,讨论了模型参数对支持向量回归机预测精度的影响,为模型参数寻优提供一般指导原则.直门达水文站径流量混沌时间序列支持向量回归机预测实验表明,水文混沌时间序列支持向量回归机模型是有效的.     

基于多尺度自适应阶ARMA的混沌序列多步预测

混沌时间序列在自然界以及人们的生产生活中很常见,混沌序列看似杂乱无章但相较于纯随机序列其中蕴含着一些非线性的运动特征,提出一种基于多尺度自适应阶ARMA的混沌时间序列多步预测方法.首先利用自适应噪声的完备经验模态分解(CEEMDAN)对原始混沌序列进行分解,获得不同尺度的固有模态分量(IMF)和残余分量.然后采用经粒子群算法(PSO)进行阶数寻优的自回归移动平均模型(ARMA)对每一个IMF分量进行拟合预测.最后将预测得到的每一个分量相加得到原始混沌序列的预测值.基于Mackay-Glass混沌序列和太阳黑子数混沌序列进行实验分析,实验表明:与ARMA、PSO-ARMA以及CEEMDAN-ARMA方法相比,方法的预测效果有较好的提高,其平均绝对误差(MAE)、均方根误差(RMSE)以及平均绝对百分比误差(MAPE)都有降低.     

CPI数据预测计算的ANFIS模型结构探讨

依据CPI经济序列数据确定性混沌原理,探讨自适应神经模糊推理系统模型构造,并给出此类混沌数据列预测的ANFIS系统结构形式,进行CPI经济序列数据预测.并用实例拟合、预测数据证明:ANFIS模型是一种精度较高的混沌数据序列预报系统.为CPI数据预测提供了一种计算方法.     

基于支持向量机的混沌时间序列模型在房产价格指数预测中的应用

针对房产价格指数的预测问题,建立了混沌时间序列的支持向量机的非线性预测模型.首先运用Cao氏法进行相空间重构,并利用改进型小数据量法计算最大的Lyapunov指数,分析上海房产价格指数时间序列的混沌特性.然后以最小嵌入维数作为支持向量机的输入节点,建立房地价格指数的预测模型.实例表明,该方法能较好地处理复杂的房地产数据,具有较高的泛化能力和很好的预测精度.     

以混沌理论为基础的神经网络预测方法

我国证券市场股价波动表现出特有的混沌性质，具有局部随机与整体秩序相容的特征。本以2002年每隔十秒的上证指数高频数据为例，以混沌理论为基础。从原始序列中构造出若干个新的时间序列，运用神经网络法进行预测。预测结果表明，此方法能够较好地预测股票的走势，有望在股票交易中应用。     

太阳辐射时间序列的非线性检验

检验太阳辐射时间序列是否有非线性特征,对于分析、建模和预测太阳辐射量是重要、有益的.提出用基于替代数据的检验方法来检验太阳辐射时间序列是否存在非线性特征,并将数据序列的三阶矩作为检验统计量.选取了美国Montana州Dillon地区和Wyoming州Green Rivet地区每日总辐射量、Utah州Moab地区的每月日平均总辐射量时间序列作为检验对象.数值分析的统计结果表明所研究的日总辐射时间序列存在非线性,而每月日平均总辐射时间序列未检测出非线性.因而,对太阳辐射时间序列建模和预测之前,检验其是否有非线性特征是必要的.     

Mathematical modeling and analysis of the depletion of dissolved oxygen in water bodies

In this paper, a nonlinear mathematical model is proposed to study the depletion of dissolved oxygen in a water body caused by industrial and household discharges of organic matters (pollutants). The problem is formulated as a food chain model by considering various interaction processes (biodegradation and biochemical) involving organic pollutants, bacteria, protozoa, an aquatic population and dissolved oxygen. Using stability theory, it is shown that as the rate of introduction of organic pollutants in a water body increases, the concentration of dissolved oxygen decreases due to various interaction processes. It is found that if the organic pollutants are continuously discharged into water body, the concentration of dissolved oxygen may become negligibly small, thus threatening the survival of aquatic populations. However, by using some effort to control the cumulative discharge of these pollutants into the water body, the concentration of dissolved oxygen can be maintained at a desired level.     

基于混沌的三江平原月降水时间序列分析

三江平原是我国最大的淡水沼泽区,近年来降水的减少是导致湿地减少的一个重要的自然因素.别拉洪河是三江平原上比较有代表性的沼泽性河流.以别拉洪水文站的降水序列为例,通过相空间重构,分别计算了序列的关联维、最大Lyapunov指数以及Kolmogorov熵等几个序列特征量.计算表明:三江平原月降水序列中是存在明显的混沌特征的,这为以后建立三江平原月降水的混沌预测模型提供了理论依据.     

Random signal analysis in an environmental sciences problem

Two major difficulties are encountered in the identification of wastewater treatment plant and river water quality dynamics: process behaviour can neither be easily observed, nor easily experimented upon, and the underlying biological nature of the processes involved is only partially understood. This paper describes the derivation of a model for the interaction between dissolved oxygen (DO), biochemical oxygen demand (BOD), and algae in a freshwater river. Noisy measurements from a stretch of the River Cam downstream of Cambridge are analysed using various techniques of identification, parameter estimation and filtering. An important feature of the paper is the interpretation of system identification as a hypothesis testing/decision making procedure.     

Fractionally differenced models for water quality time series

This paper deals with the selection and evaluation of statistical techniques for use in the modeling and forecasting of water quality time series. The focus is on statistical concepts relevant to the analysis of flows and concentrations. A selection of time series procedures has been used for auditing water quality archival data, including the screening of data sets, correlation and spectrum calculations, and iterative model fitting. A summary is provided of experience with analyzing archival data on the Niagara River and the use of a fractionally differenced model.This paper is the result of a study performed for the International Joint Commission, United States and Canada. The authors gratefully acknowledge the direction and support provided by Dr. Joel L. Fisher.     

Modeling the depletion of dissolved oxygen in a water body located near a city

Water bodies located nearby cities are much prone to pollution, especially in the developing countries, where effluents treatment facilities are generally lacking. The main reason for this phenomenon is the increasing population in the cities, and the large number of industries located near them. This leads to generation of huge amounts of domestic and industrial sewage that is discharged into the water bodies, increasing their organic pollutant load and resulting in the depletion of dissolved oxygen. In this paper, we propose a mathematical model for this situation, focusing especially on the resulting quality of the water, determined by the level of dissolved oxygen. The model also accounts for resources needed for the population survival and for the industrial operations. In addition, we describe also the decomposition of organic pollutants by bacteria in the aquatic medium. Feasibility conditions and stability criteria of the system's equilibria are determined analytically. The results show that human population and industries are relevant influential factors responsible for the increase in organic pollutants and the decrease in dissolved oxygen in the water body, in the sense that they may exert a destabilizing effect on the system. The numerical simulations confirm the analytical results. Copyright © 2016 John Wiley & Sons, Ltd.     

Management strategies for the control of eutrophication processes in Fogliano lagoon (Italy): a long-term analysis using a mathematical model

Models are developed and used to analyse and test different management strategies aimed at limiting eutrophication processes in Fogliano Lagoon: modification of lagoon hydrodynamics by tidal flow regulation, harvest of algae biomass, reclaim of sediments. Mathematical models, which have been constructed and proposed, simulate, on a multiyear time scale, the main ecological processes responsible for the most important effects of eutrophication: vegetal blooms, summer anoxia. For different management strategies, hydrodynamic fields produced by wind and tide, and three-dimensional concentration fields of significant species in the ecological phenomena, in water and into sediments, are quantified and compared. The species simulated are: in the water column dissolved oxygen, phytoplanktonic biomass, macrophytic biomass, orthophosphate, dissolved organic carbon, particulate organic carbon and hydrogen sulphide; in sediments dissolved oxygen, dissolved organic carbon, particulate organic carbon, orthophosphate, adsorbed phosphorous and hydrogen sulphide. On the basis of the results of the simulations carried out, the best management strategy limiting eutrophication processes in Fogliano lagoon has been pointed out.     

MODELING AND ANALYSIS OF THE DEPLETION OF ORGANIC POLLUTANTS BY BACTERIA WITH EXPLICIT DEPENDENCE ON DISSOLVED OXYGEN

In this paper, a nonlinear mathematical model is proposed and analyzed to study the depletion of dissolved oxygen caused by interactions of organic pollutants with bacteria in a water body, such as lake. The system is assumed to be governed by three dependent variables, namely, the cumulative concentration of organic pollutants, the density of bacteria and the concentration of dissolved oxygen. In the model, the coefficient of interaction of organic pollutants with bacteria depends upon the concentration of dissolved oxygen nonlinearly and explicitly, which is the main focus of this paper, has never been studied before. The stability theory of differential equations is used to analyze the model and to confirm the analytical results numerical simulation is performed. The model analysis shows that if the coefficient of interaction mentioned above depends upon dissolved oxygen explicitly, the decrease in its concentration is more than the case when the interaction does not depend on dissolved oxygen and consequently the depletion of organic pollutants is also more in such a case.     

Extended and Unscented Kalman filtering based feedforward neural networks for time series prediction

With the ability to deal with high non-linearity, artificial neural networks (ANNs) and support vector machines (SVMs) have been widely studied and successfully applied to time series prediction. However, good fitting results of ANNs and SVMs to nonlinear models do not guarantee an equally good prediction performance. One main reason is that their dynamics and properties are changing with time, and another key problem is the inherent noise of the fitting data. Nonlinear filtering methods have some advantages such as handling additive noises and following the movement of a system when the underlying model is evolving through time. The present paper investigates time series prediction algorithms by using a combination of nonlinear filtering approaches and the feedforward neural network (FNN). The nonlinear filtering model is established by using the FNN’s weights to present state equation and the FNN’s output to present the observation equation, and the input vector to the FNN is composed of the predicted signal with given length, then the extended Kalman filtering (EKF) and Unscented Kalman filtering (UKF) are used to online train the FNN. Time series prediction results are presented by the predicted observation value of nonlinear filtering approaches. To evaluate the proposed methods, the developed techniques are applied to the predictions of one simulated Mackey-Glass chaotic time series and one real monthly mean water levels time series. Generally, the prediction accuracy of the UKF-based FNN is better than the EKF-based FNN when the model is highly nonlinear. However, comparing from prediction accuracy and computational effort based on the prediction model proposed in our study, we draw the conclusion that the EKF-based FNN is superior to the UKF-based FNN for the theoretical Mackey-Glass time series prediction and the real monthly mean water levels time series prediction.     

MODELING AND FORECASTING MONTHLY AVERAGE WATER LEVELS BASED ON THE ELLIPTIC ORBIT MODEL

Evaluation and forecasting of water‐level fluctuation for one river is of increasing importance since it is intimately associated with human welfare and socioeconomic sustainability development. In this study, it is found that time series of monthly water‐level fluctuation exhibits annual cyclical variation. Then with annual periodic extension for monthly water‐level fluctuation, the so‐called “elliptic orbit model” is proposed for describing monthly water‐level fluctuation by mapping its time series into the polar coordinates. Experiments and result analysis indicate potentiality of the proposed method that it yields satisfying results in evaluating and forecasting monthly water‐level fluctuation at the monitoring stations in the Yangtze River of China. It is shown that the monthly water‐level fluctuation is well described by the proposed elliptic orbit model, which offers a vivid approach for modeling and forecasting monthly water‐level fluctuation in a concise and intuitive way.