基于相关系数的滑坡位移新陈代谢组合预测研究  

Study on metabolic combined prediction model of landslide displacement based on correlation coefficient

作　　者：党升 冯晓[1,2] 卢志豪 陈茂霖 韦春桃[1] DANG Sheng;FENG Xiao;LU Zhihao;CHEN Maolin;WEI Chuntao(School of Civil Engineering,Chongqing Jiaotong University,Chongqing 400074,China;Institute of Engineering Information&3S,Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]重庆交通大学土木工程学院,重庆400074 [2]重庆交通大学工程信息与3S研究所,重庆400074

出　　处：《人民长江》2021年第7期95-100,共6页Yangtze River

基　　金：国家自然科学基金资助项目(41801394);重庆市基础与前沿研究计划资助项目(cstc2015jcyjBX0023)。

摘　　要：针对目前常规组合预测模型在滑坡位移预测预报中精度下降过快的问题,从基于相关系数的组合预测模型出发,应用新陈代谢理论提出了一种基于相关系数的滑坡位移新陈代谢组合预测模型,定义、推导了相应的计算公式,并引入预测评价指标体系对该模型的预测效果进行评价。同时,以黄茨滑坡和新滩滑坡位移为实例进行了验证,结果表明:基于相关系数的滑坡位移新陈代谢组合预测模型的预测评价指标优于单项预测模型和基于相关系数的组合预测模型。Aiming at the problem that the accuracy of conventional combined prediction model in landslide displacement prediction decreases too fast,starting from the combined prediction model based on correlation coefficient,a metabolic theory was applied to propose a metabolic combined prediction model for landslide displacement based on correlation coefficient.The corresponding calculation formula were defined and deduced,and the prediction effect of the model was evaluated by introducing the prediction evaluation index system.Through two examples of Huangci landslide and Xintan landslide,the verified results showed that the prediction evaluation index of the metabolic combined prediction model for landslide displacement based on correlation coefficient was better than the single prediction model and the combined prediction model based on correlation coefficient.

关 键 词：滑坡位移 相关系数 新陈代谢法 组合预测模型 黄茨滑坡 新滩滑坡 

分 类 号：P642[天文地球—工程地质学] P258[天文地球—地质矿产勘探]