| As a kind of clean energy,geothermal energy has been paid more and more attention by scholars and government departments.The geothermal resources is mainly distributed at the edge of tectonic plates,but our country has a unique "ladder" topography,and there are many areas at the boundary of plate collision,and abundant geothermal resources.The Yinchuan Basin is a part of North China block,which lies between Qinghai-Tibet Block,Alashan block and Ordos block since Cenozoic era.The uplift and transfer effect of the Tibetan Plateau made the Ordos block slowly move to the southeast and formed the peripheral basin group after the end of the Cretaceous,which laid the basic tectonic background for the formation and accumulation of geothermal resources.The hydrothermal geothermal resources in Yinchuan Basin are conduction-type geothermal areas in sedimentary basins,which have been explored and exploited.However,the deep geothermal resource types and the genetic mechanism of deep geothermal resources are lacking,so further exploration is needed.The geophysical exploration method is an important method for deep geothermal exploration,which can provide quantitative basis.In this paper,gravity,magnetic and electrical geophysical exploration data are used to achieve high-precision inversion and imaging of underground structures,and to determine the spatial distribution characteristics of resistivity,density and magnetic susceptibility of deep rocks in Yinchuan Basin.At the same time,Bayesian thermal parameter inversion based on interface information was used to obtain the spatial distribution of mantle heat flow and other thermal parameters in the study area.Based on the comprehensive geophysical inversion results,the geothermal genetic mechanism and geothermal model of Yinchuan Basin were analyzed and established,which provided the basis for regional geothermal resource evaluation and development.The Gauss Newton method is used for 2D inversion of magnetotelluric sounding data,and the classical Occam inversion results are proposed as the initial value of Gauss Newton inversion,which improves the computational efficiency and accuracy.The low resistance block model and fault model were established for simulation calculation and inversion calculation analysis,and the reliability of the method was confirmed.The one-dimensional and two-dimensional resistivity inversion of the magnetotelluric data in Yinchuan Basin was carried out to obtain the characteristics of the underground electrical structure and determine the deep and large fault system,which can provide geological interpretation for the deep heat transfer channel.In addition,in order to overcome the problem of multiple solutions of a single geophysical method,a mencomputer interactive inversion was carried out based on the measured gravity data,and the existence of a source with high conductivity,low velocity and low density in the deep part of Yinchuan Basin was determined,which was presumed to be a hightemperature molten body,providing a heat source for the formation of regional geothermal resources.Aiming at the problems of difficult parameter selection,non-uniqueness of inversion results and low resolution of regional gravity and magnetic data joint inversion method,the gravity and magnetic joint inversion technology based on deep learning U-net network is proposed to obtain the distribution of density and magnetic susceptibility in the deep part of Yinchuan Basin with high precision,and provide basic data for the distribution of heat storage and the interpretation of regional heat control conditions.Firstly,a random complex underground geological model is established to form the gravity and magnetic data set required for network training.Then,the training data set is input into the dual channel U-net neural network,and the complex mapping relationship between gravity and magnetic data and inversion model is constructed to realize the deep fusion of multiple geophysical data.The inversion result shows that this method has good generalization and robustness.This method is used to invert gravity and magnetic anomalies in Yinchuan Basin.The results show that the upper uplift of Ordovician basement(and possibly including Tertiary strata)and local high density anomalies are formed.This feature provides important clues for delineating heat storage structures and searching for geothermal resources.The deep basement obtained by inversion has strong local magnetism,which is of great significance for the analysis of the distribution law of metamorphic rock basement.It is a good thermal conductivity layer in the deep,and can provide heat source for shallow heat reservoir under the control of deep and large faults such as the Yellow River fault,and also provides an important clue for deep dry hot rock exploration.Aiming at the problems of few well temperature measurement data and unclear understanding of regional thermal state,the sequential inversion method of various thermal parameters based on interface information is researched.In this paper,gravity and magnetic measurements are used to obtain underground Moho and interior information,and the inversion method based on MCMC(Markov chain-Montocaro)is developed to obtain heat flow,thermal conductivity,heat generation rate and other important thermal parameters.Based on the model calculation,the geothermal model of the crust and deep lithosphere of the Yinchuan Basin is reversed,and the regional heat flow,thermal conductivity and other thermal parameters are obtained,which provides reliable information for the geothermal genetic modeling and multi-type geothermal resources investigation in the Yinchuan Basin.Combined with the results of regional geological and geophysical inversion imaging,it is believed that the heat source in the deep mantle migrated along the fault to the shallow layer,forming a high heat anomaly area in the eastern part of the shallow surface,which is in contrast with the high heat anomaly area in the western part of the deep basin,providing a key basis and evidence for the formation and evolution of the deep basin heat source and its genetic mechanism.Based on the results of seismic exploration and inversion results of gravity,magnetic and magnetotelluric measured data in Yinchuan Basin,the distribution characteristics of regional geothermal resources and geothermal genetic models in Yinchuan Basin were studied,and a variety of geophysical detection methods including gravity,magnetic and electric were adopted to achieve high-precision inversion of underground structures.The spatial distribution characteristics of resistivity,density and magnetic susceptibility of underground media in Yinchuan Basin were obtained.Based on the interface information,Bayesian thermal parameter inversion was used to obtain the spatial distribution of mantle heat flow and other thermal parameters in the study area,and the geothermal genetic mechanism and geothermal genetic model of Yinchuan Basin were established.That is,Yinchuan Basin was under the effect of the remote stress field of the Tibetan Plateau,and the extensional environment enabled the upwelling of underground hot materials and uplift of the basin basement.In addition,positive faults in the tensile environment provide upwelling channels for hot materials,during which the groundwater is heated and the shallow cap layer is baked,forming the distribution of abundant geothermal resources in the eastern part of the study area.The hot dry rock resources are formed in deep metamorphic rocks and possibly hidden rock bodies,which will provide an important foundation for geothermal exploration in the future.On the basis of comprehensive geophysical inversion and analysis,the geothermal resource prospective area is delineated,which provides theoretical support and scientific basis for geothermal resource exploration and sustainable development and utilization in Yinchuan Basin. |
PDF Full Text Request