一种改进的人工鱼群聚类算法

为了克服K-Means算法对初始类簇中心、噪声点、孤立点敏感缺点,将K-Means算法和人工鱼群算法结合,提出了改进的人工鱼群聚类算法。在该算法中将类簇中心看作一条人工鱼,让每条人工鱼执行随机、觅食、聚群、追尾行为中的一种,并将更新后的位置作为K-Means算法的初始值,不断重复人工鱼的位置更新和K-Means操作,直到算法结束。由于在算法中加入了动态移动步长和全局人最优人工鱼位置,聚类的收敛精度和速度都得到提高。使用iris和glass数据集进行聚类时,与其他算法相比,文中的收敛时间缩短2.6%,精度提高1.36%。     

用对称映射ARMA模型的零极点研究子波相位对反射系数序列反演的影响

为研究地震子波相位对反射系数序列反演的影响,在自回归滑动平均(ARMA)模型描述子波的基础上,提出采用z域对称映射ARMA模型零极点的方法构造了一系列相同振幅谱、不同相位谱的地震子波,并结合谱除法对人工合成地震记录进行反射系数序列反演.理论分析表明,子波相位估计不准时反射系数序列反演结果中残留一个纯相位滤波器,该纯相位滤波器的相位谱为真实子波和构造子波的相位谱之差.采用丰度和变分作为评价方法,在反演结果中确定出真实的或准确的反射系数序列.仿真实验和实际数据处理结果也验证了子波相位对反射系数序列反演的影响规律和评价方法的有效性,为进一步提高反射系数序列反演结果精度指明了研究方向.     

Evaluating channel response to an extreme sedimentation event in the context of historical range of variability: Upper Colorado River,USA

In May 2003, a breach in a large irrigation ditch within Rocky Mountain National Park (RMNP) initiated a debris flow that entered Lulu Creek and the Colorado River, where 36 000 m³ of sediment substantially altered channel forms and processes. We present a proof of concept to understand whether the 2003 disturbance is within the historical range of variability (HRV), and whether the recovery potential of the system is sufficient to adapt to the disturbance. Flow and sediment regimes, and channel morphology and stability were monitored on Lulu Creek and the Colorado River from 2004 to 2011. Dominant channel response following the debris flow within Lulu Creek included step development, bed armoring, and channel widening. Step height‐to‐length ratios (H/L) for three reaches on Lulu Creek are outside the HRV of reference channels, with one reach approaching reference conditions. Erosion of approximately 23% of the debris fan volume occurred as a result of the long duration 2011 peak flow. Sediment within the Lulu Creek fan will persist for ~30–190 years, assuming current maximum and mean removal rates. Planform changes on the Colorado River since the debris flow include an increase in single‐thread geometries, with braided reaches where bar deposition occurred. Bedload transport and grain‐size analysis of bedload indicate translational spreading of a sand wave front with a dispersive component in steeper reaches. Lulu Creek is returning to a condition of natural variability, but the Colorado River is outside the HRV expected for steep‐gradient, pool‐riffle channels. Applying HRV to a situation where management questions require a longer term perspective, and pre‐disturbance baseline data are limited, is a useful approach. The HRV analysis facilitates a better understanding of site variability and delineates the range of possibilities of channel form and process to achieve management goals. Copyright © 2012 John Wiley & Sons, Ltd.     

Improving the performance of data-driven techniques through data pre-processing for modelling daily reservoir inflow

Abstract

The effect of data pre-processing while developing artificial intelligence (AI) -based data-driven techniques, such as artificial neural networks (ANN), model trees (MT) and linear genetic programming (LGP), is studied for Pawana Reservoir in Maharashtra, India. The daily one-step-ahead inflow forecasts are compared with flows generated from a univariate autoregressive integrated moving average (ARIMA) model. For the full-year data series, a large error is found mainly due to the occurrence of zero values, since the reservoir is located in an intermittent river. Hence, all the techniques are evaluated using two data series: 18 years of daily full-year inflow data (from 1 January to 31 December); and 18 years of daily monsoon season inflow data (from 1 June to 31 October) to take into account the intermittent nature of the data. The relevant range of inputs for each category is selected based on autocorrelation and partial autocorrelation analyses of the inflow series. Conventional pre-processing methods, such as transformation and/or normalization of data, do not perform well because of the large variation in magnitudes, as well as the many zero values (65% of the full-year data series). Therefore, the input data are pre-processed into un-weighted moving average (MA) series of 3 days, 5 days and 7 days. The 3-day MA series performs better, maintaining the peak inflow pattern as in the actual data series, while the coarser-scale (5-day and 7-day) MA series reduce the peak inflow pattern, leading to more errors in peak inflow prediction. The results indicate that AI methods are powerful tools for modelling the daily flow time series with appropriate data pre-processing, in spite of the presence of many zero values. The time-lagged recurrent network (TLRN) ANN modelling technique applied in this study maps the inflow forecasting in a better way than the standard multilayer perceptron (MLP) neural networks, especially in the case of the seasonal data series. The MT technique performs equally well for low and medium inflows, but fails to predict the peak inflows. However, LGP outperforms the other AI models, and also the ARIMA model, for all inflow magnitudes. In the LGP model, the daily full-year data series with more zero inflow values performs better than the daily seasonal models.

Citation Jothiprakash, V. & Kote, A. S. (2011) Improving the performance of data-driven techniques through data pre-processing for modelling daily reservoir inflow. Hydrol. Sci. J. 56(1), 168–186.     

Simulating sub-daily hydrological process with SWAT: a review

ABSTRACT

The Soil and Water Assessment Tool (SWAT) is a watershed-scale hydrologic model that integrates water quantity and quality modules. Despite the large amount of knowledge on the SWAT model, specific understanding of sub-daily applications remains limited. In this review, we identify the shortcomings and possible ways forward in simulating sub-daily processes with the model. A literature review was conducted, along with a participatory method based on a questionnaire. We reviewed 28 scientific articles and categorized them into: (i) model development, (ii) streamflow methods comparison, (iii) water quality, and (iv) other applications. We found that using sub-daily data improves hydrograph peak simulation, while for medium flows use of daily data was better. From all the reviewed studies, a 1-hour time step was the most suitable time scale for the sub-daily model application. The participatory questionnaire confirmed the hypothesis that the main challenge for using the sub-daily routine was the lack of high-resolution data.     

欧拉反褶积在重磁位场中应用与发展

介绍了欧拉反褶积方法原理,概述了国内外近几十年地球物理学者对欧拉反褶积方法进行的一系列改进及相关问题研究取得的主要进展和成就,并着重分析了构造指数的选取和欧拉解的稳定性等问题。通过分析指出了构造指数的正确选取、多场源混合叠加干扰和高阶次背景场研究及欧拉解的稳定性仍是今后欧拉反褶积方法研究重点和发展的主要方向,这对解决欧拉反褶积在重磁位场资料解释实用化中具有重要意义。     

二维大地电磁正演中的无网格算法

无网格法是近几年来发展的一种新的基于变分原理的数值计算方法,由于在计算形函数中不需要划分网格.在力学、电磁学等领域得到了广泛的研究.基于无网格法在大地电磁勘探正演中的应用进行了研究,首先对无网格法的基本原理进行了阐述,并利用广义变分原理推导出了相应的离散方程,编制了相应的程序,最后通过两个理论模型的计算结果检验了算法的正确性.     

An Unusual, Three-Dimensionally Preserved, Large Hadrosauriform Pes Track from “Mid”-Cretaceous Jiaguan Formation of Chongqing, China

Three-dimensional tracks provide unique insights into the locomotor mechanics of their track makers. An isolated, large hadrosauriform print attributable to Caririchnium lotus from the “mid”-Cretaceous Lotus track site (Jiaguan Formation) in China permits reconstruction of the footfall, weight-bearing, and kick-off phases of the step cycle. Large-scale modifications of the pes during the step cycle indicate C. lotus trackmakers were capable of locomotory modifications in response to substrate consistency beyond the “expected” shift between bipedal and quadrupedal postures. An unusual curvature to the trace of one of the outer digits indicates substantial transverse mobility. The remaining digits demonstrate lesser degrees of transverse movement accompanied by extension of the digits during footfall. The absence of overprinted scale-scratch marks and toe drags are consistent with a vertical kick-off of the pes and concomitant flexion of the digits. This track suggests that pedal mobility in C. lotus track makers was greater than previously suspected and has implications for reconstructions of hadrosauriform locomotion.     

斜向波对台阶地形上有限吃水半无限长弹性板的散射作用

基于线性势流理论和欧拉-伯努利梁理论,采用匹配特征函数展开法建立斜向波对台阶地形上有限吃水半无限长弹性板散射的解析解。与以往研究不同,本文同时考虑波浪入射角度、弹性板吃水、三种不同板端条件（自由板端、简支板端、固定板端）以及海底地形的影响。本文还直接将板端条件引入匹配边界条件,建立线性方程组求解速度势中的展开系数,计算过程简单、高效。计算结果表明：本文方法收敛性好,并满足能量守恒方程。将本文计算结果与文献中的特例结果进行对比,吻合良好。分析了波浪入射角度、弹性板吃水、不同板端条件以及海底地形对弹性板水动力参数的影响,研究结果可以为工程设计提供参考。     

一维全非线性Green-Naghdi水波方程数值模型

建立了求解一维全非线性Green-Naghdi水波方程的中心有限体积/有限差分混合数值格式。采用结构化网格对守恒形式的控制方程进行离散和积分,界面数值通量采用有限体积法计算,剩余项则采用中心有限差分格式求解。其中,采用中心迎风有限体积格式计算控制体界面数值通量,并结合界面变量的线性重构方法,使其在空间上具有四阶精度,通过引入静压重构技术和波浪破碎指标使模型具备处理海岸水-陆动边界及波浪破碎的能力。时间积分则采用具有总时间变差减小(Total Variation Diminishing,TVD)性质的三阶龙格-库塔法进行。应用该模型对孤立波在常水深和斜坡海岸上的传播过程及规则波跨越潜堤传播的实验进行了数值模型研究,数值计算同解析解及实验数据吻合良好。