论文部分内容阅读
针对传统滑坡预测手段数据源有限、数据更新周期长、难以发现隐藏在复杂滑坡系统中的规律等问题,本文以三峡库区为研究对象,从多源空间数据中提取滑坡孕灾环境和影响因素等信息,采用数字地形水文分析方法划分斜坡单元,对评价因子进行重采样,进而构建两类支持向量机模型。分析了多源影响因素与滑坡易发性的定量关系,并生成滑坡易发性分区图。采用成功率曲线和误差率评价预测结果,模型预测精度达到98.21%,与野外调查实际情况吻合较好。
In view of the limited data sources of traditional landslide prediction methods and the long data update period, it is difficult to find the hidden rules in complex landslide systems. Taking the Three Gorges reservoir area as the research object, this paper extracts the landslide hazard environment and its influencing factors And other information, using digital terrain hydrological analysis method to divide slope units, resampling evaluation factors, and then constructing two support vector machine models. The quantitative relationship between multi-source influencing factors and landslide susceptibility was analyzed, and the landslide susceptibility map was generated. Using the prediction of success rate curve and error rate, the prediction accuracy of the model reaches 98.21%, which is in good agreement with the field survey.