Improved Gath–Geva clustering for fuzzy segmentation of hydrometeorological time series

In this paper, an improved Gath–Geva clustering algorithm is proposed for automatic fuzzy segmentation of univariate and multivariate hydrometeorological time series. The algorithm considers time series segmentation problem as Gath–Geva clustering with the minimum message length criterion as segmentation order selection criterion. One characteristic of the improved Gath–Geva clustering algorithm is its unsupervised nature which can automatically determine the optimal segmentation order. Another characteristic is the application of the modified component-wise expectation maximization algorithm in Gath–Geva clustering which can avoid the drawbacks of the classical expectation maximization algorithm: the sensitivity to initialization and the need to avoid the boundary of the parameter space. The other characteristic is the improvement of numerical stability by integrating segmentation order selection into model parameter estimation procedure. The proposed algorithm has been experimentally tested on artificial and hydrometeorological time series. The obtained experimental results show the effectiveness of our proposed algorithm.     

Dynamic programming approach for segmentation of multivariate time series

In this paper, dynamic programming (DP) algorithm is applied to automatically segment multivariate time series. The definition and recursive formulation of segment errors of univariate time series are extended to multivariate time series, so that DP algorithm is computationally viable for multivariate time series. The order of autoregression and segmentation are simultaneously determined by Schwarz’s Bayesian information criterion. The segmentation procedure is evaluated with artificially synthesized and hydrometeorological multivariate time series. Synthetic multivariate time series are generated by threshold autoregressive model, and in real-world multivariate time series experiment we propose that besides the regression by constant, autoregression should be taken into account. The experimental studies show that the proposed algorithm performs well.     

Modified dynamic programming approach for offline segmentation of long hydrometeorological time series

For the offline segmentation of long hydrometeological time series, a new algorithm which combines the dynamic programming with the recently introduced remaining cost concept of branch-and-bound approach is developed. The algorithm is called modified dynamic programming (mDP) and segments the time series based on the first-order statistical moment. Experiments are performed to test the algorithm on both real world and artificial time series comprising of hundreds or even thousands of terms. The experiments show that the mDP algorithm produces accurate segmentations in much shorter time than previously proposed segmentation algorithms.     

Similarity search and pattern discovery in hydrological time series data mining

The rapid development of data mining provides a new method for water resource management, hydrology and hydroinformatics research. In the paper, based on data mining theory and technology, we analyse hydrological daily discharge time series of the Shaligunlanke Station in the Tarim River Basin in China from the year 1961 to 2000. Firstly, according to the four monthly statistics, namely mean monthly discharge, monthly maximum discharge, monthly amplitude and monthly standard deviation, K‐mean clustering was used to segment the annual process of the daily discharge. The clustering result showed that the annual process of the daily discharge can be divided into five segments: snowmelt period I (April), snowmelt period II (May), rainfall period I (June–August), rainfall period II (September) and dry period (October–December and January–March). Secondly, dynamic time warping (DTW), which is a different distance metric method from the traditional Euclidian distance metric, was used to look for similarities in the discharge process. On the basis of the similarity matrix, the similar discharge processes can be mined in each period. Thirdly, agglomerative hierarchical clustering was used to cluster and discover the discharge patterns in terms of the autoregressive model. It was found that the discharge had a close relationship with the temperature and the precipitation, and the discharge processes were more similar under the same climatic condition. Our study shows that data mining is a feasible and efficient approach to discover the hidden information in the historical hydrological data and mining the implicative laws under the hydrological process. Copyright © 2010 John Wiley & Sons, Ltd.     

A dynamic programming segmentation procedure for hydrological and environmental time series

We present a procedure for the segmentation of hydrological and environmental time series. The procedure is based on the minimization of Hubert’s segmentation cost or various generalizations of this cost. This is achieved through a dynamic programming algorithm, which is guaranteed to find the globally optimal segmentations with K=1, 2, ..., K _max segments. Various enhancements can be used to speed up the basic dynamic programming algorithm, for example recursive computation of segment errors and “block segmentation”. The “true” value of K is selected through the use of the Bayesian information criterion. We evaluate the segmentation procedure with experiments which involve artificial as well as temperature and river discharge time series.     

A hidden Markov model segmentation procedure for hydrological and environmental time series

In this paper we present a procedure for the segmentation of hydrological and enviromental time series. We consider the segmentation problem from a purely computational point of view which involves the minimization of Huberts segmentation cost; in addition this least squares segmentation is equivalent to Maximum Likelihood segmentation. Our segmentation procedure maximizes Likelihood and minimizes Huberts least squares criterion using a hidden Markov model (HMM) segmentation algorithm. This algorithm is guaranteed to achieve a local maximum of the Likelihood. We evaluate the segmentation procedure with numerical experiments which involve artificial, temperature and river discharge time series. In all experiments, the procedure actually achieves the global minimum of the Likelihood; furthermore execution time is only a few seconds, even for time series with over a thousand terms.     

利用动态时间规整方法实现不同时间长度地震观测资料的形态匹配

本文将动态时间规整方法引入到地震观测资料的形态匹配分析中,以解决因时间尺度不一致的两列观测数据无法定量比对的问题。基于动态时间规整技术方法原理,通过测试数据验证了动态时间规整方法的可行性,并利用云南西部地区的断层实际观测数据,分析了1996年丽江M_S7.0地震前的跨断层观测数据异常形态与当前数据异常形态的相似性问题。结果表明:① 动态时间规整方法可用于地震资料时间长度不一致时的相似性匹配;② 时间不一致的两列观测数据可用累积规整路径距离来定量表征,累积距离越短,曲线形态越一致;③ 动态时间规整方法可用于给定模板的前兆数据相似度的计算机自动提取,可提高当前仅依靠人工判别的工作效率;④ 从模式识别的角度考虑,当前下关跨断层水准观测数据变化形态与1996年丽江M_S7.0地震和2008年汶川M_S8.0地震前的水准数据变化形态较为一致。      

Segmentation algorithm for long time series analysis

Time series analysis is an important issue in the earth science-related engineering applications such as hydrology, meteorology and environmetrics. Inconsistency and nonhomogeneity that might arise in a time series yield segments with different statistical characteristics. In this study, an algorithm based on the first order statistical moment (average) of a time series is developed and applied on five time series with length ranging from 84 items to nearly 1,300. Comparison to the existing segmentation algorithms proves the applicability and usefulness of the proposed algorithm in long hydrometeorological and geophysical time series analysis.     

用逐步代价最小决策法识别地震与爆破

在动态时间规整法的基础上, 建立了逐步代价最小决策法(SAMC). 该方法中的代价函数可以很好地反映特征归属, 对较差的特征具有一定的“容忍度”、 稳定性好, 还可用全程代价函数评判识别结果的可信度. 用SAMC方法对北京及其周边地区33次地震和29次爆破中提取的5个分类特征量进行识别, 识别率为90%; 从该5个特征量中选择较好的3个特征量进行识别, 识别率为92%; 在上述地区另选13次事件作为检验样本进行U检验, 5个分类特征量和3个分类特征量的识别率分别为92%和100%, 识别效果很好. 这表明SAMC是识别地震与爆破的有效方法.      

基于递归分析和聚类的大地电磁信噪辨识及分离

为了剖析大地电磁信号和强干扰的本质特征,进一步精细分离出微弱的大地电磁有用信号,提出基于递归分析和聚类的大地电磁信噪辨识及分离方法.首先,运用递归分析法扩展大地电磁一维时间序列的维数,分析了嵌入维数、延迟时间和判别阈值对递归图的性能,并研究了不同长度的序列对递归定量分析参数的影响情况,然后,构建典型的大地电磁强干扰类型和微弱的大地电磁有用信号样本库,针对样本库讨论了强干扰和微弱大地电磁信号之间的递归定量分析参数,分析了K均值聚类和模糊C均值聚类的信噪辨识效果.最后,对实测大地电磁数据进行信噪辨识处理,并仅对辨识为强干扰的时间段采用数学形态滤波进行噪声压制.实验结果表明,递归分析能定性及定量地描述大地电磁信号时间序列的非线性特征和原动力系统的本质规律,与聚类算法相结合能对矿集区实测大地电磁信号进行信噪辨识;处理后的卡尼亚电阻率-相位曲线更为光滑、连续,其结果更为精细地保留了大地电磁信号低频段的缓变化信息,整个低频段的大地电磁数据质量得到了明显改善.     

Fast segmentation algorithms for long hydrometeorological time series

A time series with natural or artificially created inhomogeneities can be segmented into parts with different statistical characteristics. In this study, three algorithms are presented for time series segmentation; the first is based on dynamic programming and the second and the third—the latter being an improved version of the former—are based on the branch‐and‐bound approach. The algorithms divide the time series into segments using the first order statistical moment (average). Tested on real world time series of several hundred or even over a thousand terms the algorithms perform segmentation satisfactorily and fast. Copyright © 2008 John Wiley & Sons, Ltd.     

Comparison of spectral and time domain calibration methods for precipitation‐discharge processes

Hydrological model parameter estimation is an important aspect in hydrologic modelling. Usually, parameters are estimated through an objective function minimization, quantifying the mismatch between the model results and the observations. The objective function choice has a large impact on the sensitivity analysis and calibration outcomes. In this study, it is assessed whether spectral objective functions can compete with an objective function in the time domain for optimization of the Soil and Water Assessment Tool (SWAT). Three empirical spectral objective functions were applied, based on matching (i) Fourier amplitude spectra, (ii) periodograms and (iii) Fourier series of simulated and observed discharge time series. It is shown that most sensitive parameters and their optimal values are distinct for different objective functions. The best results were found through calibration with an objective function based on the square difference between the simulated and observed discharge Fourier series coefficients. The potential strengths and weaknesses of using a spectral objective function as compared to utilising a time domain objective function are discussed. Copyright © 2010 John Wiley & Sons, Ltd.     

A SVD-based ensemble projection algorithm for calculating the conditional nonlinear optimal perturbation

Conditional nonlinear optimal perturbation(CNOP) is an extension of the linear singular vector technique in the nonlinear regime.It represents the initial perturbation that is subjected to a given physical constraint,and results in the largest nonlinear evolution at the prediction time.CNOP-type errors play an important role in the predictability of weather and climate.Generally,when calculating CNOP in a complicated numerical model,we need the gradient of the objective function with respect to the initial perturbations to provide the descent direction for searching the phase space.The adjoint technique is widely used to calculate the gradient of the objective function.However,it is difficult and cumbersome to construct the adjoint model of a complicated numerical model,which imposes a limitation on the application of CNOP.Based on previous research,this study proposes a new ensemble projection algorithm based on singular vector decomposition(SVD).The new algorithm avoids the localization procedure of previous ensemble projection algorithms,and overcomes the uncertainty caused by choosing the localization radius empirically.The new algorithm is applied to calculate the CNOP in an intermediate forecasting model.The results show that the CNOP obtained by the new ensemble-based algorithm can effectively approximate that calculated by the adjoint algorithm,and retains the general spatial characteristics of the latter.Hence,the new SVD-based ensemble projection algorithm proposed in this study is an effective method of approximating the CNOP.     

区域特性约束下的油藏物性模拟

为了提高油藏物性模拟的精度,本文提出了包含两个约束条件的目标函数,考虑在既满足井条件与空间结构变化,同时又满足油田范围内观测到的油藏物理性质的约束下,应用模拟退火算法,对孔隙度的空间分布进行预测模拟.由于目标函数有效地结合了静、动态信息,从而降低了模拟结果的不确定性,提高了模型的精度.合成数据的油藏物性模拟试验表明,该方法收敛快、稳定性好,模拟的结果令人满意.     

Novel deseasonalizing models for improving the prediction of total ozone in column using evolutionary programming and neural networks

In this paper we present a novel method for deseasonalizing TOC data using non-linear models, with evolutionary computation techniques, and its performance with a neural network as regression approach. Specifically, the proposed deseasonalization method uses an evolutionary programming (EP) approach to carry out a curve fitting problem, where a given function model is optimized to be as similar as possible to an objective curve (a real TOC measurement in this case). Different non-linear models are proposed to be optimized with the EP algorithm. In addition, we test the possibility of deseasonalizing the TOC measurement and also the meteorological input data. The deseasonalized series is then used to train a neural network (multi-layer perceptron). We test the proposed models in the prediction of several TOC series in the Iberian Peninsula, where we carry out a comparison against a reference deseasonalizing model previously proposed in the literature. The results obtained show the good performance of some of the deseasonalizing models proposed in this paper.     

Modified precise time step integration method of structural dynamic analysis

The precise time step integration method proposed for linear time-invariant homogeneous dynamic systems can provide precise numerical results that approach an exact solution at the integration points. However, difficulty arises when the algorithm is used for non-homogeneous dynamic systems, due to the inverse matrix calculation and the simulation accuracy of the applied loading. By combining the Gaussian quadrature method and state space theory with the calculation technique of matrix exponential function in the precise time step integration method, a new modified precise time step integration method （e.g., an algorithm with an arbitrary order of accuracy） is proposed. In the new method, no inverse matrix calculation or simulation of the applied loading is needed, and the computing efficiency is improved. In particular, the proposed method is independent of the quality of the matrix H. If the matrix H is singular or nearly singular, the advantage of the method is remarkable. The numerical stability of the proposed algorithm is discussed and a numerical example is given to demonstrate the validity and efficiency of the algorithm.     

垂直上升管中气液两相流电导波动信号的混沌特性分析

首先以Lorenz混沌方程产生的非线性时间序列为例，讨论了在不同时间序列长度下各种延迟时间算法对噪声的适用性.研究发现，采用C_C算法计算延迟时间的鲁棒性强.在此基础上，给出了垂直上升管中气水两相流电导波动信号混沌表征结果，发现在较低水相表观速度时，随着气相表观速度增加，泡状流及混状流动力学特性变得愈加复杂，而段塞流动力学特性受液相表观速度影响较大；在较高水相表观速度时，随着气相表观速度增加，当流型从泡状流向段塞流转变时，气液两相流动力学特性变得相对简单.但是，由于受液相湍流作用影响，段塞流的动力学特性表现出了涨落现象，呈现不稳定性，当流型从段塞流向混状流转变时，气液两相流动力学特性则变得愈加复杂.研究结果表明：基于电导波动信号的混沌分析可以较好地表征气液两相流流型变化，是理解流型转变机理及其动力学演变特性的有用工具.     

Optimizing remediation of an unconfined aquifer using a hybrid algorithm

We present a novel hybrid algorithm, integrating a genetic algorithm (GA) and constrained differential dynamic programming (CDDP), to achieve remediation planning for an unconfined aquifer. The objective function includes both fixed and dynamic operation costs. GA determines the primary structure of the proposed algorithm, and a chromosome therein implemented by a series of binary digits represents a potential network design. The time-varying optimal operation cost associated with the network design is computed by the CDDP, in which is embedded a numerical transport model. Several computational approaches, including a chromosome bookkeeping procedure, are implemented to alleviate computational loading. Additionally, case studies that involve fixed and time-varying operating costs for confined and unconfined aquifers, respectively, are discussed to elucidate the effectiveness of the proposed algorithm. Simulation results indicate that the fixed costs markedly affect the optimal design, including the number and locations of the wells. Furthermore, the solution obtained using the confined approximation for an unconfined aquifer may be infeasible, as determined by an unconfined simulation.     

大样本核数量化理论Ⅳ模型及算法

把核函数理论与数量化理论IV模型有机结合,提出了核数量化理论IV模型,并以高阶对称矩阵端部特征对求解的Lanczos算法为基础设计了大样本核数量化理论IV模型的算法框架.把核数量化理论IV模型应用于高光谱图像的降维实验研究,研究结果表明,合理选择核函数模型及参数,核数量化理论IV模型能够在低维标度空间中表征原始数据的族群信息,从而取得满意的分类效果.核数量化理论IV模型为大样本地学观测数据的分析处理提供了一种新的理论工具.     

Dynamic neuro-fuzzy local modeling system with a nonlinear feature extraction for the online adaptive warning system of river temperature affected by waste cooling water discharge

This paper proposes a dynamic modeling methodology based on a dynamic neuro-fuzzy local modeling system (DNFLMS) with a nonlinear feature extraction technique for an online dynamic modeling task. Prior to model building, a nonlinear feature extraction technique called the Gamma test (GT) is proposed to compute the lowest mean squared error (MSE) that can be achieved and the quantity of data required to obtain a reliable model. Two different DNFLMS modes are developed: (1) an online one-pass clustering and the extended Kalman filtering algorithm (mode 1); and (2) hybrid learning algorithm (mode 2) of extended Kalman filtering algorithm with a back-propagation algorithm trained to the estimated MSE and number of data points determined by a nonlinear feature extraction technique. The proposed modeling methodology is applied to develop an online dynamic prediction system of river temperature to waste cooling water discharge at 1?km downstream from a thermal power station from real-time to time ahead (2?h) sequentially at the new arrival of each item of river, hydrological, meteorological, power station operational data. It is demonstrated that the DNFLMS modes 1 and 2 shows a better prediction performance and less computation time required, compared to a well-known adaptive neural-fuzzy inference system (ANFIS) and a multi-layer perceptron (MLP) trained with the back propagation (BP) learning algorithm, due to local generalization approach and one-pass learning algorithm implemented in the DNFLMS. It is shown that the DNFLMS mode 1 is that it can be used for an online modeling task without a large amount of training set required by the off-line learning algorithm of MLP-BP and ANFIS. The integration of the DNFLMS mode 2 with a nonlinear feature extraction technique shows that it can improve model generalization capability and reduce model development time by eliminating iterative procedures of model construction using a stopping criterion in training and the quantity of required available data in training given by the GT.