Font Size: a A A

Study On Seismic Wave Field Characteristic Analysis And Waveform Inversion Method Of Coal Seam

Posted on:2020-07-30Degree:DoctorType:Dissertation
Country:ChinaCandidate:P WangFull Text:PDF
GTID:1360330572480620Subject:Earth Exploration and Information Technology
Abstract/Summary:PDF Full Text Request
As one of the most significant energy structure in China,coal are and still will play an indispensible part in the social and economic development in the future.In recent years,coalfield exploration has gradually emphasized the deep coal seams and complex tectonic areas,which puts forward increasingly higher requirements for fine seismic processing and interpretation technology.At the same time,the realization of fine seismic processing and interpretation technology depends on the efficient use of seismic full-wave field information.However,the current coalfield seismic exploration still focuses on the use of one-time reflection wave inversion method to obtain coal seam-related medium parameters.Therefore,it is urgent to develop a full wave inversion method for coalfields.Being a high-precision quantitative seismic inversion method,full-waveform inversion is based on wave equation,which makes full advantage of the information of seismic full-wave field to obtain formation medium parameters in depth domain.As a result,it will become an extremely important development trend of coalfield exploration,which is precisely the jumping-off point of this paper.The concept of full waveform inversion was put forward in the 1980s.Unfortunately,some drawbacks exist in it,such as the large amount of calculation and poor stability.Until recent years,full waveform inversion method has begun to develop rapidly in the field of petroleum exploration with the continuous improvement of computer hardware and software performance.At present,it is widely applied to the model testing and offshore data.Compared with oil and gas reservoirs,the coal seams has the following characteristics.First and foremost,coal seams are buried shallow and the coalfield seismic data are susceptible to noise interference,which leads to poor quality of seismic data.Secondly,with thin and low-speed beds,coal seams often exist in the form of multiple sets of coal interbeds,which results in more complex seismic waveform response law of coal seams than oil and gas reservoirs.Next,coal measures strata are destroyed and collapse pillars are formed due to the karst collapse.It is more difficult to operate the inversion characterization of the special geological structure of the small-scale coal seam than that of the conventional fault system in oil and gas fields.Consequently,the current full waveform inversion methods and technical strategies,which are designed to oil and gas exploration,cannot be directly applied,when the full waveform inversion technology is used in coalfield seismic exploration.Therefore,the main problem to be solved in this paper is to study the law of seismic wave propagation and the corresponding full waveform inversion technology for coal seams.Initial inversion model is vital in full waveform inversion,which is often obtained by seismic structural interpretation and log data interpolation.Its correctness determines the convergence speed and stability of full waveform inversion iteration.In full waveform inversion of elastic wave in coalfield,the establishment of initial model requires not only conventional P-wave logging data,but also shear-wave logging data.Actually,there is less shear wave logging data due to cost problems in actual production and shear-wave velocity is often fitted by empirical relationship,which results in large errors.In order to improve the accuracy of modeling,the elastic and anisotropic parameters of representative coal seams are obtained from rock physics testing in this paper.Based on artificial intelligence algorithm,the reconstruction model of S-wave velocity logging curve based on conventional logging parameters is established,and its correctness is verified.A new method of estimating S-wave velocity based on artificial intelligence is proposed innovatively,which offers seismic wave field analysis for coal fields.All those lay a solid foundation for full waveform inversion of elastic wave.The strategy of full waveform inversion of coalfield seismic data depends on the forward simulation analysis of coalfield seismic wave field.As a consequence,based on the staggered grid finite difference forward simulation method of viscoelastic anisotropic wave equation,the seismic wave field response characteristics of coal seam thickness,viscoelasticity,anisotropy,interbedding and collapse column geological structure are comprehensively analyzed in this paper.The research results show that the strong reflection waveform of coal seam is formed by the reflection of the top and bottom interface.In the case of multiple sets of coal seams,the law of tuning influence is more complex.Although the time domain waveform can not effectively distinguish coal seams,the characteristics of coal seams can be reflected at different degrees in dissimilar frequency scales.Under the circumstance of certain coal seam thickness,the influence of viscoelasticity and anisotropy on seismic reflection wave field of coal seams can be neglected.Geological structures of typical coalfields such as collapse pillars,can make seismic wave field more complex due to the diffraction of faults or breakpoints.These laws and understandings provide a direction for the follow-up study of the full waveform inversion method of coalfield seismic data.Considering the difference of frequency response of reflection waveform tuning law in thin interbedded strata,the multi-scale envelope-waveform inversion method for thin interbedded geological structure is studied in this paper.As for the complexity of seismic wave field generated by fault diffraction wave of collapse column in coalfield,a stable convergence time domain full waveform inversion method based on inexact Newton optimization algorithm is proposed in that the coalseam thickness is buried deeply and it is difficult to estimate its thickness.A regularized full waveform inversion method with well constraints is developed in this paper.Take the characteristics of coal measures strata in coalfield into consideration,the above three technical methods are integrated to form a combined optimization seismic full waveform inversion method for coalfield exploration.The results of Model tests show that this method can effectively invert the collapse column structure and coal seam structure,which gives technical support to seismic inversion of coal seams,as well as lays a solid foundation for the full waveform inversion of actual seismic data in coal field in the future.
Keywords/Search Tags:coalfield seismic exploration, artificial intelligence, full waveform inversion of elastic waves in time domain, inexact newton, multiscale
PDF Full Text Request
Related items