地球物理学进展 ›› 2021, Vol. 36 ›› Issue (4): 1521-1530.doi: 10.6038/pg2021EE0315

• 应用地球物理学Ⅰ(油气及金属矿产地球物理勘探) • 上一篇    下一篇

圆锥型场源瞬变电磁数据AWPSO算法优化反演

黄刚1,2,3(), 杨海燕3,4,*(), 余国锋1,2, 李锋平4, 彭峰3, 焦俊俊3, 刘建鹏5, 张华3, 顾艳艳3   

  1. 1.平安煤炭开采工程技术研究院有限责任公司,淮南 232000
    2.深部煤炭开采与环境保护国家重点实验室,淮南 232000
    3.东华理工大学地球物理与测控技术学院,南昌 330013
    4.中国矿业大学资源与地球科学学院,徐州 221116
    5.核工业二九〇研究所,韶关 512029
  • 收稿日期:2020-10-17 修回日期:2021-02-18 出版日期:2021-08-20 发布日期:2021-09-15
  • 通讯作者: 杨海燕 E-mail:gangh16@163.com;genious_yang@126.com
  • 作者简介:黄刚,男,1994年生,硕士研究生,主要从事瞬变电磁法勘探理论与应用研究. E-mail: gangh16@163.com
  • 基金资助:
    国家自然科学基金(41974086);安徽省科技重大专项(17030901023);东华理工大学研究生创新基金联合资助(YC2020-S484);东华理工大学研究生创新基金联合资助(DHYC-202011)

Optimal inversion of transient electromagnetic data with a conical source excitation based on AWPSO algorithm

HUANG Gang1,2,3(), YANG HaiYan3,4,*(), YU GuoFeng1,2, LI FengPing4, PENG Feng3, JIAO JunJun3, LIU JianPeng5, ZHANG Hua3, GU YanYan3   

  1. 1. Ping’an Coal Mining Engineering Technology Research Institute Co., Ltd., Huainan 232000, China
    2. State Key Laboratory of Deep Coal Mining and Environment Protection, Coal Mining National Engineering Technology Research Institute, Huainan 23200, China
    3. School of Geophysics and Measurement and Control Technology, East China University of Technology, Nanchang 330013, China
    4. School of Resources and Geosciences, China University of Mining and Technology, Xuzhou 221116, China
    5. Institute 290 of Nuclear Industry, Shaoguan 512029, China
  • Received:2020-10-17 Revised:2021-02-18 Online:2021-08-20 Published:2021-09-15
  • Contact: YANG HaiYan E-mail:gangh16@163.com;genious_yang@126.com

摘要:

瞬变电磁圆锥型场源装置有效的减小了线圈间的电感,提高了小装置探测地下浅层的分辨率,但常规反演方法需给定初始模型且反演精度不高.针对瞬变电磁法反演计算问题,通过对粒子群优化算法(PSO)和神经网络算法(BP)分析研究,改进了一种基于神经网络算法Sigmoid函数的自适应加权粒子群优化(AWPSO)算法.采用标准测试函数对算法进行试算,建立多个理论层状地质模型对该算法进行理论验证,最后在地质资料已知地区开展现场实验.计算结果表明,新提出的算法具有更高的全局搜索寻优能力和收敛速度快、计算精度高,且不需要给定初始模型;实验结果显示实测数据反演结果与高密度电法探测结果吻合,证明该算法能够对瞬变电磁探测数据进行反演计算且精度较高,可以在同类型的浅层探测任务中提供参考.

关键词: 瞬变电磁法, 圆锥型场源, PSO算法, AWPSO算法, 高密度电法, 反演

Abstract:

Transient electromagnetic conical source device can effectively reduce the inductance between coils and improve the resolution of detecting the underground shallow layer of the small device. However, the conventional inversion method requires a given initial model and the inversion accuracy is not high.To solve the problem of transient electromagnetic inversion, an Adaptive Weighted Particle Swarm Optimization (AWPSO) algorithm based on neural network algorithm Sigmoid function was improved by analyzing and studying PSO and BP.The standard test function was used to test the algorithm, and multiple theoretical layered geological models were established to verify the algorithm. Finally, field experiments were carried out in areas with known geological data.The results show that the new algorithm has higher global search optimization ability, faster convergence speed and higher calculation accuracy, and does not need the initial model.The experimental results show that the inversion results of the measured data are consistent with the high-density electric method detection results, which proves that the algorithm can carry out inversion calculation of the tem data with high accuracy, and can provide reference for the same type of shallow detection tasks.

Key words: Transient electromagnetic method, Conical source, Particle Swarm Optimization(PSO), Adaptive Weighted Particle Swarm Optimization Algorithm(AWPSO), High-density electrical method, Inversion

中图分类号: