航空重力梯度的模拟退火参数反演  

Inversion of airborne gravity gradient by using simulated annealing algorithm

作　　者：刘金钊[1] 王岩 高春春 李永昆[1] 王勇 LIU J inzhao;WANG Yan;GAO Chunchun;LI Yongkun;WANG Yong(The First Monito-ring and Application Center,China Earthquake Administration,Tianjin 300180,China;Nanyang Normal University,Nanyang,Henan 473061,China;School of Geology and Geomatics,Tianjin Chengjian University,Tianjin 300384,China)

机构地区：[1]中国地震局第一监测中心,天津300180 [2]南阳师范学院,河南南阳473061 [3]天津城建大学地质与测绘学院,天津300384

出　　处：《测绘科学》2018年第8期7-13,共7页Science of Surveying and Mapping

基　　金：中国地震局星火计划项目(XH17058Y);中国地震局第一监测中心主任创新基金项目(FMC2016004);天津市自然科学基金项目(17JCYBJC21600)

摘　　要：针对目前地球物理问题反演时,通常利用线性或拟线性反演方法,从而导致反演过程不稳定或反演结果陷于局部最优的问题,该文采用了模拟退火算法的非线性全局反演技术,从航空重力梯度全张量数据中反演了地下密度异常体的几何参数信息。同时,利用无噪声和有噪声的两组航空重力梯度数据进行对比实验。结果表明,在重力梯度数据信噪比小于21.584 3的情况下,通过引入噪声加权的价值函数,可以提高地下密度异常体几何参数反演的准确性。Aiming at the problem that traditional linear inversion or quasi linear inversion methods on geophysical data can bring unstable inversion process and get local optimum results,this paper implemented geometry parameters inversion of underground anomaly from the airborne gravity gradient full tensor data by using simulated annealing algorithm which is global nonlinear inversion technique.Meanwhile,comparison experiments were carried out for two groups of airborne gravity gradient data with and without noise respectively.The results demonstrated that when the signal-to-noise ratio of gravity gradient data was less than 21.584 3,the accuracy of the parameters inversion of underground anomaly could be improved by introducing the weighted noise cost function.

关 键 词：重力梯度 模拟退火算法 参数反演 加权 

分 类 号：P223[天文地球—大地测量学与测量工程]