摘要
岚皋县因为独特的地貌和多变的气候,地质灾害相对较多。为了减少灾害所带来的经济损失,建立岚皋县地质灾害预警的BP神经网络细网格模型,对该地区地质灾害进行预测分析。选取地形地貌,地层岩性,地形坡度,降雨作为评价指标并运用GIS技术将该地区地质图剖分成368个2.5(km)*2.5(km)的细网格。在此基础上,计算每个网格的降雨等级,致灾因素综合叠加后的危险性等级以及灾害等级作为模型的训练数据。使用该数据训练后,对岚皋县大道河镇地质灾害进行预测,发现预测结果与实际情况基本一致。研究表明,使用该模型进行地质灾害预测较为可行。
Due to its unique landform and variable climate,Langao country has relatively more geological disasters.In order to reduce the economic losses caused by disasters,a fine mesh model of the BP neural network for the early warning of geological disasters in Langao country is established,and geological hazards in the region are predicted and analyzed.Topographical features,formation lithology,topographic gradient,and rainfall are selected as evaluation indicators and GIS techniques are used to map the region’s geological map into 3682.5(km)*2.5(km)fine grids.On this basis,the rainfall level of each grid is calculated,and the hazard level and disaster level after the comprehensive addition of the hazards are used as training data for the model.After using this data to train and predict the geological disasters in Dadaohe town,Langao county,it is found that the forecast results are basically consistent with the actual situation.Studies have shown that using this model for geological disaster prediction is more feasible.
作者
杨昆
朱志祥
YANG Kun;ZHU Zhixiang(Institute of Internet of Things and IT-based Industrialization,Xi'an University of Post&Telecommunications,Xi'an 710061)
出处
《计算机与数字工程》
2019年第12期3107-3114,共8页
Computer & Digital Engineering
基金
陕西省重点研发计划项目(编号:2016KTTSGY01-01)
西安邮电大学教学改革研究项目(编号:JGZ201615)资助
关键词
岚皋县
细网格预警
地质灾害
BP神经网络
GIS技术
Langao country
fine-srid forecast
geological hazard
BP neural network
GIS technology