Study on magnetotelluric inversion method based on hybrid intelligent algorithm optimization

被引：0
|
作者：
Wang X. [1 ]
Bai N. [1 ]
Zhou J. [2 ]
Wang Y. [3 ]
机构：
[1] School Of Physics And Electronic Information Engineering, He Nan Polytechnic University, Jiaozuo
[2] Institute Of Geophysics And Geomatics, China University Of Geosciences, Wuhan
[3] Henan Jinghui Technology Co., Ltd., Zhengzhou
来源：
Meitan Kexue Jishu/Coal Science and Technology (Peking) | 2021年 / 49卷 / 07期
关键词：
Differential evolution algorithm; Nelder-mead optimization algorithm; Particle swarm optimization algorithm; Robustness;
D O I：
10.13199/j.cnki.cst.2021.07.020
中图分类号：
学科分类号：
摘要：
The inversion problem of magnetotelluric is non-linear. If the linear inversion method is used to linearize the non-linear problem, it is easy to fall into the local minimum and the non uniqueness of the inversion results is serious. At present,the single nonlinear intelligent algorithm is easy to fall into local minimum and low efficiency in inversion. In order to solve this problem,this paper designs a nonlinear hybrid intelligent algorithm combining particle swarm optimization,differential evolution algorithm and Nelder-Mead optimization algorithm. At the same time,the hybrid intelligent algorithm makes use of the ability of particle swarm optimization and differential evolution algorithm for global optimization. At the same time,it also uses the Nelder-Mead optimization algorithm for further local development of the optimization results, which solves the problems of linear algorithm and single nonlinear intelligent algorithm easily falling into local minimum and low efficiency of optimization. In the process of magnetotelluric inversion, regularization theory is introduced to constrain the objective function to avoid non-uniqueness of the inversion results. The damped particle swarm optimization, differential evolution and the hybrid intelligent algorithm proposed in this paper are respectively applied to the typical multidimensional function optimization and theoretical model. The results show that the hybrid intelligent algorithm in this paper has high efficiency and good robustness compared with damped particle swarm optimization algorithm and differential evolution algorithm. Finally,the hybrid intelligent algorithm is applied to the inversion of a measured point data of Jilin huapichang geothermal field,and the inversion results are compared with the inversion results of the traditional inversion algorithm, which shows that the inversion results of the algorithm in this paper are more consistent with the actual exploration results, which proves that the hybrid intelligent algorithm has a high accuracy in the inversion of magnetotelluric measured data. © 2021 China Coal Society. All Rights Reserved.
引用
收藏
页码：147 / 153
页数：6
相关论文
共 20 条
  • [1] TANG Jingtian, REN Zhengyong, ZHOU Cong, Et al., Frequency - domain electromagnetic methods for exploration of the shallow sub? surface:a review [ J], Chinese Journal of Geophysics, 58, 8, pp. 2681-2705, (2015)
  • [2] YE Gaofeng, WANG Hui, GUO Zeqiu, Et al., Data acquisition and processing technology of long - period magnetotclluric [ J ], Progressin Geophys, 28, 3, pp. 1219-1226, (2013)
  • [3] CHEN Li, QIN Qiming, WANG Nan, Review of the forward model? ing and inversion in magnetotelluric sounding field[ J], Acta Sci? entiarum Naturalium Universitatis Pekinensis, 50, 5, pp. 979-984, (2014)
  • [4] CONSTABLE S C, PARKER R L, CONSTABLE C G., Occam's in? version:a practical algorithm for generating smooth models from EM sounding data, Geophysics, 52, 3, pp. 289-300, (1987)
  • [5] FENG Sichen, WANG Xuben, RUAN Shuai, Comparison among several inversion algorithms of 1D MT[ J], Oil Geophysical Pros? pecting, 39, 5, pp. 594-599, (2004)
  • [6] LIU Yunfeng, CAO Chunlei, Inversion of one-dimensional MT data using genetic algorithms[J], Journal of Zhejiang University, 31, 3, pp. 300-305, (1997)
  • [7] SHI Xueming, WANG Jiaying, One dimensional magnetotelluric sounding inversion using simulated anneaking, Earth Science: Journal of China University of Geoscience, 23, 5, pp. 543-545, (1998)
  • [8] SHI Xueming, XIAO Min, FAN Jianke, Et al., The damped PSO al? gorithm and its application for magnetotelluric sounding data inver? sion [ J ], Chinese Journal of Geophysics, 52, 4, pp. 1114-1120, (2009)
  • [9] WANG Tianyi, YANG Jin, YAN Tingjie, Et al., The differential evolution algorithm in geophysical inversion [ J ], Geology and Prospecting, 50, 5, pp. 971-975, (2014)
  • [10] XIONG Jie, Member IAENG, Et al., A non-linear geophysical in? version algorithm for the MT data based on improved differential evolution[J], Engineering Letters, 26, 1, pp. 161-170, (2018)
12