摘要
陷落柱对煤矿安全生产威胁极大,矿井TEM和地震勘探是2种主要的导水陷落柱探查方法,矿井TEM对富水性敏感,地震勘探空间分辨率高,为了充分利用2种方法各自优势,提出了基于矿井TEM和地震联合反演的导水陷落柱精细探测方法。首先,提出了基于镜像2D电阻率模型和全空间1D TEM正演算法的矿井拟二维(Pseudo-2D)TEM反演算法。矿井Pseudo-2D TEM反演算法既满足交叉梯度联合反演算法对2D模型的需要,又具有较快的反演计算速度。然后,通过去掉常规联合反演目标函数中的地震正演项,只在交叉梯度项中保留地震波阻抗模型参数,建立了单向交叉梯度联合反演算法。在单向交叉梯度联合反演算法中,地震波阻抗模型只对TEM反演进行约束,电阻率模型不对地震反演进行约束,地震波阻抗反演可以通过商业软件完成,TEM反演采用矿井Pseudo-2D TEM反演算法。通过对地震波阻抗模型进行双三次插值处理,实现地震波阻抗模型和电阻率模型之间的模型网格匹配。通过基于K-means算法的地震波阻抗模型聚类分割处理方法消除地震波阻抗模型中的次要结构变化,增加联合反演稳定性。最后,建立了含导水陷落柱地质-地球物理模型,采用有限差分方法分别进行了3D矿井TEM和3D地震勘探正演计算,对正演模拟数据进行了单独反演和联合反演计算。结果表明:联合反演能够同时重建理论模型的形状和电性特征,联合反演结果不仅具有较高的空间分辨率,而且能够反映陷落柱的富水性,优于单独TEM和单独地震方法反演结果。相比传统联合反演方法主要应用于大尺度目标体探测或浅地表成像,基于矿井TEM和地震的单向交叉梯度联合反演方法可以实现大深度小尺度目标体的精细探测。
Karstic collapse column seriously threatens mine safety.Mine TEM and seismic exploration are two key widely applied methods for a practical karstic collapse column detection.Mine TEM is sensitive to water abundance and seismic method has high spatial resolution.To utilize the advantage of each method,the joint inversion of mine TEM and seismic exploration is proposed.First,based on the mirror 2D resistivity model and the 1D TEM whole space forward algorithm,the mine Pseudo-2D TEM inversion algorithm is proposed.The mine Pseudo-2D TEM inversion algorithm fulfills the joint inversion requirements of 2D model and cost-effective calculation time.Then,through removing the seismic forward term from the conventional joint inversion objective function(only preserving the seismic impedance model parameters in the cross gradient term),a unidirectional cross-gradient joint inversion algorithm is formulated.The seismic impedance model constrains the TEM inversion while the resistivity model does not constrain the seismic inversion in the unidi-rectional cross-gradient joint inversion.Seismic impedance inversion can be done by commercial software,and TEM inversion is based on the mine Pseudo-2D TEM inversion algorithm.Furthermore,the interpolation of seismic impedance model is proposed to achieve the match of seismic and TEM model grids.The clustering and segmentation of seismic impedance model based on K-means algorithm is proposed,which eliminates the secondary structure changes in the seismic impedance model and increases the stability of joint inversion.Finally,a model with a water conducted karstic collapse column is built,a 3D mine TEM simulation and a 3D seismic simulation are carried out.Independent inversion and joint inversion are calculated.Results show that joint inversion recovers both the shape and the electrical characteristics of the theoretical model.Joint inversion results not only have a high spatial resolution but also detect water abundance,which is better than independent TEM and seismic inversion results.Different from the traditional joint inversion method mainly used in a large scale target detection or shallow surface imaging,the unidirectional cross-gradient joint inversion based on mine TEM and seismic is reliable for the detection of targets with small scale and large depth.
作者
李飞
程久龙
杨思通
温来福
董毅
LI Fei;CHENG Jiulong;YANG Sitong;WEN Laifu;DONG Yi(State Key Laboratory of Coal Resources and Safe Mining,China University of Mining and Technology (Beijing),Beijing 100083,China;Hebei Key Laboratory of Mine Disaster Prevention and Control,North China Institute of Science and Technology,Beijing 101601,China;College of Earth Science and Engineering,Shandong University of Science and Technology,Qingdao 266590,China)
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2020年第7期2472-2481,共10页
Journal of China Coal Society
基金
国家重点研发计划资助项目(2017YFC0804105)
教育部创新团队资助项目(IRT_17R37)
河北省自然科学基金资助项目(D2019508160)。
关键词
矿井瞬变电磁法
地震勘探
联合反演
导水陷落柱
交叉梯度
mine TEM
seismic exploration
joint inversion
water conducted karstic collapse column
cross-gradient