Identification of physical processes inherent in artificial neural network rainfall runoff models

The emergence of artificial neural network (ANN) technology has provided many promising results in the field of hydrology and water resources simulation. However, one of the major criticisms of ANN hydrologic models is that they do not consider/explain the underlying physical processes in a watershed, resulting in them being labelled as black‐box models. This paper discusses a research study conducted in order to examine whether or not the physical processes in a watershed are inherent in a trained ANN rainfall‐runoff model. The investigation is based on analysing definite statistical measures of strength of relationship between the disintegrated hidden neuron responses of an ANN model and its input variables, as well as various deterministic components of a conceptual rainfall‐runoff model. The approach is illustrated by presenting a case study for the Kentucky River watershed. The results suggest that the distributed structure of the ANN is able to capture certain physical behaviour of the rainfall‐runoff process. The results demonstrate that the hidden neurons in the ANN rainfall‐runoff model approximate various components of the hydrologic system, such as infiltration, base flow, and delayed and quick surface flow, etc., and represent the rising limb and different portions of the falling limb of a flow hydrograph. Copyright © 2004 John Wiley & Sons, Ltd.     

Automatic Determination of Stellar Atmospheric Parameters Using Neural Networks and Instance-Based Machine Learning

In this article we show how machine learning methods can beeffectively applied to the problem of automatically predictingstellar atmospheric parameters from spectral information, a veryimportant problem in stellar astronomy. We apply feedforwardneural networks, Kohonen's self-organizing maps andlocally-weighted regression to predict the stellar atmosphericparameters effective temperature, surface gravity and metallicityfrom spectral indices. Our experimental results show that thethree methods are capable of predicting the parameters with verygood accuracy. Locally weighted regression gives slightly betterresults than the other methods using the original dataset asinput, while self-organizing maps outperform the other methods when significant amounts of noise are added. We also implemented a heterogeneous ensemble of predictors, combining the results given by the three algorithms. This ensemble yields better results than any of the three algorithms alone, using both the original and the noisy data.     

Reliability assessment of serviceability performance of braced retaining walls using a neural network approach

In urban environments, one major concern with deep excavations in soft clay is the potentially large ground deformations in and around the excavation. Excessive movements can damage adjacent buildings and utilities. There are many uncertainties associated with the calculation of the ultimate or serviceability performance of a braced excavation system. These include the variabilities of the loadings, geotechnical soil properties, and engineering and geometrical properties of the wall. A risk‐based approach to serviceability performance failure is necessary to incorporate systematically the uncertainties associated with the various design parameters. This paper demonstrates the use of an integrated neural network–reliability method to assess the risk of serviceability failure through the calculation of the reliability index. By first performing a series of parametric studies using the finite element method and then approximating the non‐linear limit state surface (the boundary separating the safe and ‘failure’ domains) through a neural network model, the reliability index can be determined with the aid of a spreadsheet. Two illustrative examples are presented to show how the serviceability performance for braced excavation problems can be assessed using the reliability index. Copyright © 2005 John Wiley & Sons, Ltd.     

Water and power networks and urban fragmentation in Los Angeles: Rethinking assumed mechanisms

Los Angeles is often described as the epitome of urban fragmentation, a notion which in this context is frequently connected to, or even conflated with urban sprawl. At the same time, the city features integrated water and power networks which have been under public ownership for over 70 years. We thus have an apparent paradox in the context of the debate on ‘splintering urbanism’, between socio-spatial fragmentation and the integration of networks. In discussing the idea that deregulation of infrastructural networks exacerbates urban fragmentation, the authors use the case of Los Angeles in order to highlight the central role of private interests in management decisions concerning infrastructure networks. The authors carry out their analysis in an historical perspective, revealing that network integration and universal access can often serve private interests more than the public good. Urban fragmentation in Los Angeles, they conclude, is the result of a complex process of instrumentalisation of network development and management.     

Church-based social capital, networks and geographical scale: Katrina evacuation, relocation, and recovery in a New Orleans Vietnamese American community

This research examines the role of social capital and networks to explain the evacuation, relocation, and recovery experiences of a Vietnamese American community in New Orleans, Louisiana in the aftermath of Hurricane Katrina. As the single largest community institution, the parish church’s complex bonding and bridging social capital and networks proved particularly critical in part because of its historically based ontological security. The process of evacuation, but especially relocation and recovery, was dependent on deploying co-ethnic social capital and networks at a variety of geographical scales. Beyond the local or community scale, extra-local, regional, and national scales of social capital and networks reproduced a spatially redefined Vietnamese American community. Part of the recovery process included constructing discursive place-based collective-action frames to successfully contest a nearby landfill that in turn engendered social capital and networks crossing ethnic boundaries to include the extra-local African American community. Engaging social capital and networks beyond the local geographical scale cultivated a Vietnamese American community with an emergent post-Katrina cultural and political identity.     

Identifying China’s leading world city: a network approach

This paper reports our research on China’s world cities. Formal network analysis of air passenger linkages for recent years among China’s most populous cities and among many of the world’s largest cities allows us to identify the country’s leading world city from among the leading Mainland candidates, Beijing, Shanghai and Guangzhou. We theorize our findings about China’s world cities in relation to both global forces (and China’s increasing entanglement with them) and the policies and actions of the national state. We examine the national and global urban network through a longitudinal, two-level analysis of airline passenger travel for four time points between about 1990 and 2005. We show that Beijing was China’s leading world city at the beginning of the time period, a status it lost nationally in as early as 1995, and then globally 10 years later. On the other hand Shanghai became China’s leading world city, and it acquired this status first nationally in 2000, and then globally in 2005. The changing status of the Chinese capital corresponds to the country’s increasing involvement with the capitalist world economy. Shanghai’s ascendance as the leading world city in China may indicate that global forces have come to play an increasingly important role relative to that of the developmental state.

PSO-RBFNN模型及其在岩土工程非线性时间序列预测中的应用

岩土工程受力变形演化是一个典型的非线性问题,其演化的高度非线性和复杂性,很难用简单的力学、数学模型描述,但可用粒子群优化径向基神经网络对岩土工程应力、位移非线性时间序列进行动态实时预测。网络径向基层的单元数通过均值聚类法确定后,所有其它参数：中心位置、形状参数、网络权值,均通过粒子群优化算法在全局空间优化确定。工程实例应用表明,该模型预测结果准确、精度高,有良好的应用前景。     

基于AGA的SVM需水预测模型研究

需水预测是一个由城市人口、工业水平、社会经济水平共同作用的多因素、多层次的复杂非线性系统.其结果将直接影响受区域水资源承载力约束的产业结构、布局形态等决策.作为一种集中参数预报方法,支持向量机方法具有对未来样本的较好的泛化性能,对于这类资料缺乏、系统结构尚欠清晰的问题可以取得较好的模拟和预测结果.基于此,本文将支持向量机方法引入需水预测领域,建立了需水预测支持向量机模型.同时,本文将加速遗传算法和支持向量机方法耦合起来,构造了支持向量机模型参数的自适应优化算法.模型在珠海市的应用实例表明:与简单遗传算法比较,AGA的模型参数寻优效率更高;与BP神经网络模型相比,SVM模型较好地解决了小样本、经验性等问题,并取得了较高的预测精度.     

地形坡度对多金属结核分布的控制作用

选取我国东太平洋多金属结核开辟区内的一个区域作为研究区,利用人工神经网络中应用最为广泛的BP网络,建立控制多金属结核分布的地质因素与多金属结核分布之间的映射关系,探讨地形坡度对多金属结核分布的控制作用.结果表明,在经度、纬度、水深、坡度四个因素中,坡度对结核分布影响程度最大;多金属结核主要分布于坡度小于5°的地方;当坡度小于5°时,丰度与地形坡度呈正相关,品位与地形坡度呈负相关,丰度与品位呈负相关;品位与坡度似呈指数关系,坡度增大,品位降低.     

Kohonen网络与BP网络的集成应用研究

本文介绍了 Kohonen神经网络对输入数据进行聚类方法在卷烟配方中的应用 ,提出了从核心样本动态搜索 BP网络训练样本的新探索 ,摒弃了过去 BP算法中训练样本固定不变 ,互不相交的方法 ,实现了 BP网络和 Kohonen网络动态无缝集成。