Research On Fine Detection Of Water Conducted Karstic Collapse Column Based On Joint Inversion Of Seismic And TEM

Mine water disaster caused by watercoure(karstic collapse column,fault,fractured zone and goaf)takes place constantly in recent years,which causes serious losses to people's life and property.However,for the watercourse has relatively small scale and large depth,the detection method has limitations,the fine detection of watercoure has not been realized.Fine detection requires not only accurate locate the watercourse position,but also detect the water abundance.Joint inversion has obvious advantages in improving detection accuracy and resolution.Seismic exploration has high spatial resolution and transient electromagnetic method is sensitive to water abundance.Therefore,study effective theroy and method of joint inversion in water disaster prevention and control,acomplish the fine detection of watercourse,has important theoretical and practial significance.Take the study of water conducted karstic collapse column as an example,the paper proposed unidirectional cross gridient joint inversion method based on TEM and seismic impedance.The effectiveness of unidirectional cross-gradient joint inversion method based on TEM and seismic impedance is verified by synthetic experiments and field test.The following are the main research contents,methods and conclusions.(1)Through theory analysis and formula derivation,the forward algorithms of 3D TEM,1D TEM and homogeneous medium TEM in both whole and half space are presented.?The transient electromagnetic field equation and its 3D finite difference discrete formation are derived.The model is divided into air part and earth part which simplifies the ground boundary condition.The primary and secondary field is calculated iteratively based on the equivalent surface current source.?The paper derived the frequency domain response formula of 1D TEM field in both whole and half space,accomplished the numerical calculation of frequency response by filter coefficient method and transformed the frequency response to time domain through segment approaching method.?Double model method for 3D TEM modeling based on the combination of 3D FDTD method and digital filtering method is proposed.Simulation results show that double model method can achieve high precision results with relatively small grids quantity,which decreases calculation time significantly.Double model method can achieve higher precision results compared with traditional FDTD method in the condition of using same grids quantity.(2)Through theory analysis and formula derivation,an accumulation all-time apparent resistivity is defined and the calculation formula is derived.For the resistivity calculation of one sampling time considered the result of prior sampling time,the proposed method have higher precision and resolution compared with traditional method.Besides,the paper derived the least square inversion method using linear space theory.Based on the least square inversion method,a pseudo-2D TEM inversion algorithm in both half and whole space is proposed.The pseudo-2D TEM inversion algorithm is based on 1D forward code and 2D resistivity model,which fulfills the joint inversion requirements of 2D model and cost-effective calculation time.(3)Through theory analysis and formula derivation,the unidirectional cross-gradient joint inversion method based on TEM(including mine TEM)and seismic impedance is proposed.The proposed joint inversion algorithm is composed of pseudo-2D TEM inversion algorithm,seismic impedance inversion method,unidirectional cross-gradient joint inversion algorithm,interpolation and clustering segmentation of seismic impedance model.The unidirectional cross-gradient joint inversion algorithm is formulated by removing the seismic forward term from the conventional joint inversion objective function(only preserving the seismic impedance model parameters in the cross gradient term).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 seismic impedance model and increases the stability of joint inversion.The proposed joint inversion algorithm realizes the joint inversion of TEM(including mine TEM)and multiple cover reflection seismic data and it is reliable for the detection of targets with small scale and large depth.(4)The paper built the model of North-china-type coalfields with a water conducted karstic collapse column.3D TEM forward calculation based on large fixed source observation system and 3D sicsmic forward calculation based on multiple coverage observation system were carried out.TEM inversion,impedance inversion and joint inversion of synthetic data were carried out,results show that:?impedance inversion results have high spatial resolution;?TEM inversion results have low spatial resolution to the strata and karstic collapse column;?joint inversion results have high accuracy and resolution to strata interface and collapse column boundary,and can depress the secondary field distortion caused by deviation of middle loop position.Joint inversion results not only have high spatial resolution but also detect water abundance,its detection accuracy is significantly improved compared with TEM inversion.(5)Through theory analysis,numerical simulation and field test,the paper proposed"double offset mutual inductance elimination method" and derived the correction formula,which lays the foundation of joint inversion in practice.The paper built the model of North-china-type coalfields with a water conducted karstic collapse column 30m under the coal floor.3D TEM forward calculation based on mine TEM observation system and 3D seismic forward calculation based on multiple coverage observation system is carried out.TEM inversion,impedance inversion and joint inversion of synthetic data were carried out,results show that:?impedance inversion results has high spatial resolution,but cannot detect the water abundance;?TEM inversion results cannot identify the strata interface and collapse column boundary;?Joint inversion results can accurately determine the shape and position of the collapse column,reflect the water abundance of the collapse column,and realize the fine detection of the water conducted karstic collapse column.(6)The proposed joint inversion method was successfully used in a coalmine in Ordos City to demonstrate its practical usefulness.Reprocessing of seismic data and TEM data acquisition are carried out in the test area.TEM inversion,impedance inversion and joint inversion of field data were carried out.The joint inversion results show that the buried depth of the collapse column is about 360m and the diameter of the collapse column is about 50m.The joint inversion results are consistent with the drilling results.Joint inversion results not only have high spatial resolution but also reflect the water abundance.This performance verifies the effectiveness of the proposed joint inversion method.