四川旱涝震灾害的人工神经网络外推预测

选取旱、涝、震灾害之间的一步转移概率作为Ｂ－Ｐ人工神经网络训练样本的输入信息，建立了四川旱、涝、震的人工神经网络灾型预测模型。Ｂ－Ｐ网络模型应用于实例预测结果与用主分量分析法的趋势预测结果精度接近。     

Traffic Assignment Forecast Model Research in ITS

As an important role in the development of ITS, traffic assignment forecast is always the research focus. Based on the analysis of classic traffic assignment forecast models, an improved traffic assignment forecast model, multi-ways probability and capacity constraint (MPCC) is presented. Using the new traffic as- signment forecast model to forecast the traffic volume will improve the rationality and veracity of traffic as- signment forecast.     

基于RBF神经网络的遥感影像分类算法研究

提出了运用径向基函数神经网络和K均值法进行遥感影像分类的算法，以实际的遥感影像分类为例，通过与传统的最小距离法进行比较，对RBF神经网络分类器的优点进行了归纳，并就算法实施中的一些问题进行了探讨。分析结果表明，RBF神经网络是一种有效的图像分类器。     

模糊神经网络在GPS高程转换中的应用

介绍了T-S模糊神经网络的基本原理以及如何确定GPS高程转换的模糊神经网络模型,并采用该模型对实测数据进行了计算分析。结果表明,模糊神经网络能够对小区域GPS高程做出比较准确定的拟合,从而能够为GPS高程转换提供一种较好的方法,能够满足实际工程需要。     

The Key Supply Source of Long-Distance Moisture Transport for the Extreme Rainfall Event on July 21, 2012 in Beijing

In this study, the Weather Research and Forecasting (WRF) model and meteorological observation data were used to research the long-distance moisture transport supply source of the extreme rainfall event that occurred on July 21, 2012 in Beijing. Recording a maximum rainfall amount of 460 mm in 24 h, this rainstorm event had two dominant moisture transport channels. In the early stage of the rainstorm, the first channel comprised southwesterly monsoonal moisture from the Bay of Bengal (BOB) that was directly transported to north China along the eastern edge of Tibetan Plateau (TP) by orographic uplift. During the rainstorm, the southwesterly moisture transport was weakened by the transfer of Typhoon Vicente. Moreover, the southeasterly moisture transport between the typhoon and western Pacific subtropical high (WPSH) became another dominant moisture transport channel. The moisture in the lower troposphere was mainly associated with the southeasterly moisture transport from the South China Sea and the East China Sea, and the moisture in the middle troposphere was mainly transported from the BOB and Indian Ocean. The control experiment well reproduced the distribution and intensity of rainfall and moisture transport. By comparing the control and three sensitivity experiments, we found that the moisture transported from Typhoon Vicente and a tropical cyclone in the BOB both significantly affected this extreme rainfall event. After Typhoon Vicente was removed in a sensitivity experiment, the maximum 24-h accumulated rainfall in north China was reduced by approximately 50% compared with that of the control experiment, while the rainfall after removing the tropical cyclone was reduced by 30%. When both the typhoon and tropical cyclone were removed, the southwesterly moisture transport was enhanced. Moreover, the sensitivity experiment of removing Typhoon Vicente also weakened the tropical cyclone in the BOB. Thus, the moisture pump driven by Typhoon Vicente played an important role in maintaining and strengthening the tropical cyclone in the BOB through its westerly airflow. Typhoon Vicente was not only the moisture transfer source for the southwesterly monsoonal moisture but also affected the tropical cyclone in the BOB, which was a key supply source of long-distance moisture transport for the extreme rainfall event on July 21, 2012 in Beijing.     

基于可公度性网络分析的淮河蚌埠站大洪水预测研究

根据1916—2010年淮河蚌埠水文站大洪水资料,运用信息预测理论,构建大洪水可公度有序网络结构,同时采用峰谷定位法、前兆法等方法对淮河大洪水进行综合分析和预测.结果表明:2013—2014年淮河(蚌埠站)将有可能发生大洪水.     

陕西中南部的视应力场研究

利用西安数字地震台网的宽频带数字地震记录，计算了陕西中、南部地震的视应力水平，并讨论了泾阳5．2级地震前后地震视应力的时间变化特征及视应力与震级的关系。     

A comparison of wetness indices for the prediction of observed connected saturated areas under contrasting conditions

For lack of other widely available spatial information, topography is often used to predict water fluxes and water quality in mesoscale watersheds. Such data have however proven to be misleading in many environments where large and flat valley bottoms and/or highly conducive soil covers determine water storage and water transport mechanisms. Also, the focus is generally on the prediction of saturation areas regardless of whether they are connected to the catchment hydrographic network or rather present in isolated topographic depressions. Here soil information was coupled with terrain data towards the targeted prediction of connected saturated areas. The focus was on the 30 km² Girnock catchment (Cairngorm Mountains, northeast Scotland) and its 3 km² sub‐catchment, Bruntland Burn in which seven field surveys were done to capture actual maps of connected saturated areas in both dry and humid conditions. The 1 km² resolution UK Hydrology of Soil Types (HOST) classification was used to extract relevant, spatially variable, soil parameters. Results show that connected saturated areas were fairly well predicted by wetness indices but only in wet conditions when they covered more than 30% of the whole catchment area. Geomorphic indices including information on terrain shape, steepness, aspect, soil texture and soil depth showed potential but generally performed poorly. Indices based on soil and topographic data did not have more predictive power than those based on topographic information only: this was attributed to the coarse resolution of the HOST classification. Nevertheless, analyses provided interesting insights into the scale‐dependent water storage and transport mechanisms in both study catchments. Copyright © 2013 John Wiley & Sons, Ltd.     

基于地震属性的煤层厚度预测模型及其应用

地震属性技术在岩性和构造解释等方面得到了越来越广泛的应用,特别是在煤、油气资源勘探中具有重要的作用.基于淮南矿区谢桥1区13-1煤层地震勘探资料,提取了28种地震属性数据;通过地震属性的分析,优选出平均峰值振幅、振幅的峰态、最大绝对振幅、瞬时频率斜率等4种地震属性作为煤层厚度预测模型基本参数,结合已知钻孔资料,利用多元多项式回归以及BP人工神经网络方法,求出了各属性与煤厚之间的多元多项式回归模型及人工神经网络模型,并对模型进行了误差分析和应用结果对比分析,反映出人工神经网络模型在煤厚预测中具有好的应用效果.     

Spatial rainfall model using a pattern classifier for estimating missing daily rainfall data

Missing data in daily rainfall records are very common in water engineering practice. However, they must be replaced by proper estimates to be reliably used in hydrologic models. Presented herein is an effort to develop a new spatial daily rainfall model that is specifically intended to fill in gaps in a daily rainfall dataset. The proposed model is different from a convectional daily rainfall generation scheme in that it takes advantage of concurrent measurements at the nearby sites to increase the accuracy of estimation. The model is based on a two-step approach to handle the occurrence and the amount of daily rainfalls separately. This study tested four neural network classifiers for a rainfall occurrence processor, and two regression techniques for a rainfall amount processor. The test results revealed that a probabilistic neural network approach is preferred for determining the occurrence of daily rainfalls, and a stepwise regression with a log-transformation is recommended for estimating daily rainfall amounts.