基于遗传算法优化信息量法的地灾易发性评价

Geological Hazard Susceptibility Evaluation Based on Genetic Algorithm to Optimize Information Method

为提高地质灾害风险调查评价中地质灾害易发性评价的精度,在加权信息量模型基础上,引入了遗传算法对权重进行分析,从灾害因子的等级划分和权重确定两个方面建立了完善的地质灾害易发性评价理论体系。并以定远县为研究区,在野外踏勘的基础上,选取10个指标构建地质灾害易发性评价指标体系,进行地质灾害易发性论证。结果表明,将定远县划分为极低、低、中、高、极高5个易发区,分别占全区总面积的82.6%、8.7%、5.1%、3.1%、0.5%,易发性评价分区结果与研究区野外踏勘实际情况相近。定远县地质灾害高易发区主要集中在地层岩性较软弱,高程相对较高,坡度差相对较大的地区。通过接收灵敏度曲线检验结果为97.6%,说明模型评价精度高,分区结果合理。

Aiming to improve the accuracy of geological hazard susceptibility evaluation in the investigation and evaluation of geological hazard risk,in this paper,Genetic Algorithm was introduced to analyze the weight based on the weighted information model,and a perfect theoretical system of geological hazard susceptibility evaluation was established from two aspects for the grading of hazard factors and the determination of the weight.Moreover,taking Dingyuan County as a study area,10 indicators were selected to construct the evaluation index system of geological hazard susceptibility and the susceptibility was demonstrated through field survey.The results have showed that Dingyuan County is divided into five areas with susceptibility:extremely low,low,medium,high andextremely high,accounting for 82.6%,8.7%,5.1%,3.1%and 0.5%for the total area of the whole region,respectively,and the zoning results of susceptibility evaluation are similar to the actual situation of field survey in the study area.Moreover,geological hazard susceptible area in Dingyuan County are mainly concentrated in areas with weak formation lithology,relatively high elevation and relatively large gradient difference.The test result of the receiving sensitivity curve,indicating that the evaluation accuracy of the model is high and the zoning results are reasonable.