【摘要】 为研究山洪地质灾害的智能评估与预警,以亚热带典型地区中国广西壮族自治区钟山县为研究对象,以遥感影像和实地调查为数据源,在ENVI和ArcGIS平台上处理遥感影像、光谱数据和DEM数据,全方位获取研究区的地形坡度、植被覆盖指数、土壤松散系数、山谷山脊类别、降雨量等数据。量化数据作为输入因子,以山洪灾害风险等级为输出因子,建立钟山县山洪地质灾害风险等级评价的广义回归神经网络模型。模型经过历史数据训练后,具有较强的自学习功能,通过实例验算,模型计算出的风险等级与实际风险等级吻合较好,所建模型适用于钟山县全境的山洪地质灾害风险等级评价。通过无线传输技术输入GPS定位的经纬度到模型中,自动匹配该点的神经网络输入数据,再经模型运算输出灾害风险等级,在用户终端上输出警示信息,从而实现对山洪地质灾害的实时智能预警,在钟山县具有较好的应用效果。更多还原

【Abstract】 To study the intelligent assessment and early warning of mountain flood geological disaster,Zhongshan County of Guangxi Zhuang Autonomous Region was taken as the research sample.Remote sensing images and actual surveys were used as data sources.Remote sensing images,spectral data and DEM data were processed on ENVI and ArcGIS platforms.The data of the study area were obtained,such as slope,NDVI,soil looseness coefficient,valley and ridge classification and rainfall.These data were taken as the input factors,the risk degree of the mountain flood geological disaster was taken as the output factor.A generalized regression neural network model for risk assessment of mountain flood geological disaster in Zhongshan County was established.After the training of historical data,the model has a strong self-learning function.By case checking,the risk degree calculated by the model is in good agreement with the actual risk degree.The model can be applied to the assessment of the risk degree of the mountain flood geological disaster in Zhongshan County.Through the wireless transmission technology to enter GPS positioning latitude and longitude to the model,the model platform can receive and automatically match the neural network input data,and output the disaster risk degree by the model operation,and output the warning on the user′s terminal,so as to realize the real-time intelligent early warning of the mountain flood geological disaster,which has a good application effect in Zhongshan County.更多还原