摘要
在进行地质灾害易发性分区时,常采用层次分析法确定评价因子的权重,但由于人工主观干预打分影响,造成评价因子权重的准确率与相关度不够客观全面。以蕲春县为例,通过MATLAB软件建立BP神经网络,使用机器学习算法得到评价因子最优拟合权重值后,导入ArcGIS进行空间分析,得出地质灾害易发性分区结果。对比基于层次分析法的地质灾害易发性分区结果,认为基于BP神经网络的地质灾害易发性分区结果可靠且准确度更高。BP神经网络法可有效应用于地质灾害风险评价,对防灾减灾工作具有重要意义。
In the analysis of geological disaster susceptibility,the analytic hierarchy process is often used to determine the evaluation factors weight.However,due to the influence of artificial subjective scoring,the accuracy and correlation of the evaluation factors weight are not objective and comprehensive.Taking Qichun County as an example,the BP neural network is established by MATLAB,and the optimal fitting weight of the evaluation factors is obtained by using the machine learning algorithm.Then,it is imported into ArcGIS for spatial analysis,and the results of geological disaster susceptibility division are obtained.Compared with the results of geological disaster susceptibility division based on analytic hierarchy process,it shows that the results of geological disaster susceptibility division based on BP neural network are reliable and accurate.BP neural network method can be effectively applied to geological disaster risk assessment,which is of great significance to disaster prevention and mitigation.
作者
朱文慧
邹浩
何明明
王纪云
Zhu Wenhui;Zou Hao;He Mingming;Wang Jiyun(Third Geological Brigade of Hubei Geological Bureau,Huanggang,Hubei 438000;China University of Geosciences (Wuhan),Wuhan,Hubei 430074)
出处
《资源环境与工程》
2021年第6期840-844,共5页
Resources Environment & Engineering
基金
黄冈市地质灾害精细化气象风险预警
黄冈市降雨型滑坡风险预警区划与判据研究。
关键词
BP神经网络
层次分析法
地质灾害
易发性分区
蕲春县
BP neural network
analytical hierarchy process
geological disaster
susceptibility division
Qichun County
