环境场条件对雷暴传播运动影响实例分析

通过对一个明显后向传播雷暴和一个无明显传播特征雷暴的环境场进行对比,分析环境场条件对雷暴传播运动的影响。结果表明：二者高空均受冷涡后部西北气流控制,有中空急流,低层受暖温度脊影响,气温较高,傍晚前后受短波槽影响,在鲁西北地区产生对流天气;后向传播雷暴的环境场水汽条件较好,大气斜压特征明显,近地面层高温高湿,θse锋区位于对流层中层,中层干空气与低层冷空气入侵,二者共同作用是雷暴的产生机制;无明显传播特征雷暴的环境场水汽条件较差,θse锋区位于850 hPa以下,对流层低层干冷空气与暖湿空气交绥是雷暴的产生机制;雷暴易发生在水汽通量散度中心北侧梯度较大的区域,主回波后部大气为不稳定层结且具有辐合中心、相对湿度较大的特征,这是产生新对流单体的关键;若雷暴区有湿平流,雷暴的下游方向有水汽辐合中心,且辐合中心具有斜压特征,有利于雷暴新生,反之,则不利于雷暴新生。     

建筑物震陷预测新方法研究

利用人工神经网络的基本原理，本文修正了经典BP型神经网络的激励函数，并对学习率和训练样本进行了动态调整等多方面改进。根据70个多层建筑震陷的实测资料，在分析了建筑物震陷的影响因素基础上，提取了9个指标；采用改进后的BP算法，建立了多指标的建筑物震陷预测模型。研究结果表明，改进的BP网络性能良好，所建立的模型预测精度高，具有一定的工程实用价值；神经网络法是一种有效可行的预测新方法，人工神经网络技术具有广泛的应用前景。     

三峡库区泄滩滑坡非饱和渗流分析及渗透系数反演

泄滩滑坡的水文地质结构比较复杂,滑体、滑带及滑动影响带为多孔连续介质,透水层基岩为裂隙介质。采用等效连续介质模型模拟裂隙岩体,建立了泄滩滑坡饱和－非饱和渗流的有限元计算模型。然后采用BP网络对非饱和有限元渗流计算进行了模拟,并根据蓄水期水库水位及滑坡体地下水位监测资料,采用遗传算法对泄滩滑坡各地层渗透系数进行了反演。反演的滑体饱和渗透系数为4.89?10?2 cm/s,与现场原位试验值1.78?10?2 3.2?10?2 cm/s相当;反演的滑带饱和渗透系数为4.66?10?5 cm/s,远比室内试验值2.74?10?7 5.73?10?7 cm/s大。这种差异在一定程度上反映了土体参数的尺度效应。     

GEO-Environmental Suitability Evaluation of Land for Urban Construction Based on A Back-Propagation Neural Network and GIS: A Case Study of Hangzhou

Evaluating the geo-environmental suitability of land for urban construction is an important step in the analysis of urban land use potential. Using geo-environmental factors and the land use status of Hangzhou, China, a back-propagation (BP) neural network model for the evaluation of the geo-environmental suitability of land for urban construction was established with a geographic information system (GIS) and techniques of grid, geospatial, and BP neural network analysis. Four factor groups, comprising nine separate subfactors of geo-environmental features, were selected for the model: geomorphic type, slope, site soil type, stratum steadiness, Holocene saturated soft soil depth, groundwater abundance, groundwater salinization, geologic hazard type, and geologic hazard degree. With the support of the model, the geo-environmental suitability of Hangzhou land for urban construction was divided into four suitability zones: zone I, suitable for super high-rise and high-rise buildings; zone II, suitable for multi-story buildings; zone III, suitable for low-rise buildings; and zone IV, not suitable for buildings. The results showed that a BP neural network can capture the complex non-linear relationships between the evaluation factors and the suitability level, and these results will support scientific decision-making for urban-construction land planning, management, and rational land use in Hangzhou.     

Landslide susceptibility evaluation based on BPNN and GIS: a case of Guojiaba in the Three Gorges Reservoir Area

A landslide susceptibility evaluation is vital for disaster management and development planning in the Yangtze River Three Gorges Reservoir Area. In this study, with the support of remote sensing and Geographic Information System, 4 factor groups comprising 10 separate subfactors of landslide-related data layers were selected to establish a susceptibility evaluation model based on the back-propagation neural network including slope, aspect, plan curvature, strata and lithology, distance to faults, land use/land cover, Normalized Difference Vegetation Index, Normalized Difference Water Index, distance from roads, and effect of rivers. During model development, a three-layered interconnected neural network structure of 10 (input layer) × 20 (hidden layer) × 1 (output layer) was used for evaluating the landslide susceptibility in Guojiaba. At the same time, a back-propagation algorithm was applied to calculate the weights between the input layer and the hidden layer and between the hidden layer and the output layer. The results showed that the effect of slope has the highest weight value (0.2051), which is more than two times that of the other factors, followed by strata and lithology (0.1213) and then the effect of rivers (0.1201). At the end of the susceptibility evaluation, the area was divided into four zones such as very high, high, moderate and low susceptibility. For verification, the receiver operating characteristic curve for the back-propagation neural network-derived landslide susceptibility evaluation model was drawn, and the results showed that the area under the receiver operating characteristic curve was 0.8790 and the prediction accuracy was 88%. Furthermore, the results obtained from this article were then verified by comparing with the existing landslide historical data and multiple field-verified results. Lastly, the landslide susceptibility map will help decision makers in risk management, site selection, site planning, and the design of control engineering.     

����������ε������編Ԥ��

?????????編?????????????????????????????編???????????????????????????????????????????????????????????????????????????С???????     

Resilient capacity assessment for geological failure areas: examples from communities affected by debris flow disaster

This study establishes a novel method for assessing the community resilient capacity of debris flow disasters with appropriate parameters, such as responding, monitoring and communication capabilities. This study adopts eight communities in Taiwan, namely Nangang, Tongfu, Jhongyang, Laiyuan, Chingfu, Sinsheng, Shangan and Jyunkeng, as examples. First, the Analytic Hierarchy Process was applied to establish the framework of the community resiliency capacity, including the community’s resources for disaster resilience and resident capabilities. The community’s resources for disaster resilience are identified by surveying the community leaders via checklists. Resident capabilities are determined using questionnaires. The community resilient capacity refers to the sum of the results from these two investigations. The two investigations have similar weights, indicating that they are equally significant when evaluating community resilient capacity. Second, FLO-2D software is utilized for hazard analysis by simulation results of deposited areas for debris flows, and then these areas were categorized according to hazard degrees. Finally, the vulnerability of communities is classified based on the land use type. In summary, the values of capacity, hazard and vulnerability are integrated to determine the risk of debris flow for each community. A risk map is then generated.     

基于BP神经网络的土壤冰结温度及未冰水含量预测模型

土壤冻结温度与未冻水含量是冻土的重要物理参数，影响因素多，关系复杂，利用BP网络模型来描述冻结温度与未冻水含量及其与主要影响因素之间的关系，效果良好，该模型直接根据试验数据通过神经网络的自学习能力寻求输出变量与输入变量间的内在非线性规律，其优点在于可利用一个神经网络同时描述多个因素对冻结温度及未冻水含量的影响。     

人工神经网络原理在建筑物震陷预测中的应用

运用人工神经网络原理，对BP型神经网络作了多方面的改进，采用改进后的BP算法，建立了建筑物震陷预测模型，研究结果表明，改进的BP网络性能良好，所建立的模型预测精度高，能满足工程要求，是一种有效可行的预测新方法。     

径流长期预报的人工神经网络方法

提出径流长期预报的人工神经网络方法。运用神经网络的一典型模型──“反向传播”模型,以大伙房水库在补水期的径流状况作为研究对象,尝试了神经网络方法的效果。结果表明,该方法预报成功率较高,容错能力较强,可望成为径流长期预报的有效的辅助手段。