43 results about "Velocity inversion" patented technology

The invention discloses an SAR sea surface flow velocity inversion method based on physical guidance and data driving fusion, and relates to the field of flow velocity inversion, and the method comprises the steps: carrying out the non-geophysical Doppler frequency shift correction of the Doppler frequency shift observed by a spaceborne synthetic aperture radar, and obtaining the geophysical Doppler frequency shift; predicting based on a wave-induced Doppler frequency shift model according to a radar incident angle, reanalyzed wind direction data, a sea surface 10-meter height wind speed inverted by an SAR (Synthetic Aperture Radar), a simulated wind wave parameter, a simulated surge parameter and a simulated maximum wave parameter to obtain a wave-induced Doppler frequency shift; the wave-induced Doppler frequency shift model is obtained based on physical guidance and data-driven modeling; doppler frequency shift generated by sea surface flow is determined according to the geophysical Doppler frequency shift and the wave-induced Doppler frequency shift; according to the method, the Doppler frequency shift generated based on the sea surface flow is converted to obtain the sea surface radial flow velocity, the wave-induced Doppler frequency shift is accurately removed, and reliable inversion of the sea surface flow velocity is realized.

The invention is suitable for the technical field of geophysical exploration, and provides a rock physics-driven time-frequency domain longitudinal wave velocity dispersion attribute prediction method, which comprises the following steps of: establishing a rock physics model; under the driving of the rock physical model, inverting the porosity of the microfractures in the well, calculating the skeleton elastic modulus of the rock containing the fractures, and predicting a frequency-dependent longitudinal wave velocity logging curve; constructing a time-frequency domain longitudinal wave velocity seismic inversion initial model; constructing a TF-AVO theoretical formula, developing a time-frequency domain longitudinal wave velocity seismic inversion method based on an L-BFGS algorithm, and performing longitudinal wave velocity inversion by taking the initial model as prior information; analyzing the relative change of the frequency-varying longitudinal wave speed under different frequencies, and quantitatively predicting the frequency dispersion attribute of the longitudinal wave speed; and on the basis of logging data, calibrating a longitudinal wave velocity dispersion attribute prediction result, and carrying out reservoir oil and gas detection. The method has higher precision and resolution, the stability and reliability of longitudinal wave velocity dispersion attribute prediction are remarkably improved, and the dependence on strong amplitude response is overcome.

A method for training travel time-based networks and building an image of a velocity model includes obtaining a seismic dataset of seismic traces and determining an observed travel time for each seismic trace. The method further includes obtaining a velocity network, that depends on one or more velocity parameters, and a travel time network, that depends on one or more travel time parameters. The method further includes training the velocity network and the travel time network using a cost function and an optimizer. The cost function is based on the travel times parameters, the velocity parameters, a travel times equation, a derivative of the travel times equation, and a travel time mismatch between a first observed travel time and a first travel time value output by the travel time network. The method further includes building the image of a velocity model using the trained velocity network.

A method for training travel time-based networks and building an image of a velocity model includes obtaining a seismic dataset of seismic traces and determining an observed travel time for each seismic trace. The method further includes obtaining a velocity network, that depends on one or more velocity parameters, and a travel time network, that depends on one or more travel time parameters. The method further includes training the velocity network and the travel time network using a cost function and an optimizer. The cost function is based on the travel times parameters, the velocity parameters, a travel times equation, a derivative of the travel times equation, and a travel time mismatch between a first observed travel time and a first travel time value output by the travel time network. The method further includes building the image of a velocity model using the trained velocity network.

This disclosure relates to the field of seismic exploration technology, and particularly to a method and apparatus for automatically acquiring reflection travel time for layer-by-layer tomographic velocity modeling. The method includes: acquiring a seismic time-averaged velocity function constructed from raw seismic data and a well-logging time-averaged velocity function constructed from raw well-logging data; calibrating the seismic time-averaged velocity function using the initial seismic average velocity and well-logging average velocity to obtain a calibrated seismic average velocity model; performing time-depth conversion using the calibrated seismic average velocity model to obtain calibrated depth-domain interpretation horizons; performing stacking velocity inversion layer by layer on the calibrated depth-domain horizons to obtain optimized depth-domain layer velocities; obtaining observation travel time using the stacking velocity as a criterion; and performing tomographic inversion using the calibrated depth-domain interpretation horizons, optimized depth-domain layer velocities, observation travel time, and CMP observation system to construct a depth-domain layer velocity model.

The application belongs to the technical field of oil exploration, and discloses a method for improving the precision of a velocity model by using azimuth information, which comprises the following steps: improving the signal-to-noise ratio of shot gather data; establishing an initial velocity model in the depth domain; improving the signal-to-noise ratio of pre-stack depth migration profiles; outputting reflection angle gathers; picking up azimuth residual depth differences; obtaining azimuth residual depth differences after ray tracing; obtaining new velocities and velocity updating amounts; obtaining final velocities; and performing depth migration processing, and iteratively updating the new velocities until the reflection angle gathers and common offset gathers are flattened. The application improves the velocity inversion precision of middle and deep layers and effectively improves the velocity inversion precision of shallow layers. The application is suitable for establishing a velocity model in the depth domain.

This invention provides a true surface seismic-geological integrated velocity modeling and pre-stack depth imaging method, belonging to the field of seismic imaging technology. The method includes: obtaining a near-surface velocity model using first-arrival tomographic inversion and fusing it with a mid-deep velocity model obtained from a large floating surface of reflected waves to form an initial depth-migrated velocity field; optimizing the velocity model using reflected wave tomographic velocity inversion to construct a true surface isotropic velocity model; extracting azimuth and dip data from an isotropic migration seismic database, and then calculating various parameters of the anisotropic model and obtaining anisotropic velocities through anisotropic model analysis; and repeatedly iterating and updating the model using network tomographic inversion to obtain a multi-information-constrained anisotropic velocity model. This invention solves the problems of slow modeling speed, poor accuracy, poor imaging quality, and low precision in existing true surface seismic-geological velocity modeling and pre-stack depth imaging methods.

The invention discloses a bidirectional blocking seamless fusion first arrival wave tomography inversion method and related equipment, and belongs to the technical field of petroleum and natural gas seismic prospecting. The method comprises the following steps: firstly, editing and sorting an initial speed model by using a bidirectional blocking principle to distribute first arrival time data to each block, and outputting speed model data at the same time; then, carrying out tomographic inversion block by block according to speed model data, and finally fusing speed inversion results, so as to output a fused speed model to complete a first-motion wave tomographic inversion process; according to the method, the bidirectional blocking principle is adopted, blocking is carried out in two directions at the same time, a certain span range is ensured, the problems of reducing the boundary effect and improving the inversion efficiency are considered, the complexity of large-scale data processing can be effectively reduced, and the calculation efficiency is improved; by adopting the method, the precision and reliability of inversion are improved; by adopting the method, the inversion precision and efficiency are effectively improved.

The embodiment of the invention provides a shear wave velocity inversion method and device, electronic equipment and a storage medium, and relates to the technical field of seismic wave inversion. The method comprises the following steps: acquiring a plurality of frequency dispersion energy diagrams from different seismological observation systems in the same observation area, preprocessing the plurality of frequency dispersion energy diagrams, and constructing an image matrix; constructing a target kernel matrix by using the image matrix and a sub-sample set obtained from the image matrix; performing eigenvalue decomposition on the target kernel matrix, and performing image reconstruction according to the obtained maximum eigenvalue of the target kernel matrix to obtain a fusion frequency dispersion energy diagram with higher resolution and higher quality; and performing shear wave velocity inversion by using the fused frequency dispersion energy diagram to obtain a shear wave velocity profile of the observation area, thereby remarkably improving the accuracy and reliability of the shear wave velocity inversion.

The invention relates to the technical field of image processing and deep learning, in particular to an automatic surface wave inversion method and system, and the method comprises the steps: receiving original seismic information, sequentially carrying out the mean value removal, trend removal and spectrum whitening preprocessing of surface wave information, calculating a single-component noise cross-correlation function and a nine-component noise cross-correlation function based on the preprocessed data, and obtaining a single-component noise cross-correlation function and a nine-component noise cross-correlation function; and based on a time-frequency analysis method, calculating a surface wave group velocity and a velocity frequency dispersion energy diagram through the cross-correlation function, automatically extracting a frequency dispersion curve from the velocity frequency dispersion energy diagram by using an artificial intelligence frequency dispersion extraction method, screening frequency dispersion files of which effective frequency dispersion points and signal-to-noise ratios meet a preset standard, and calculating the surface wave group velocity and the velocity frequency dispersion energy diagram through the cross-correlation function. And inverting the parameters of the three-dimensional shear wave velocity model based on the underground interface constraint three-dimensional shear wave velocity inversion method to obtain a three-dimensional shear wave velocity model. The problems that in the prior art, efficiency is low, the operation process is complex, and the professional threshold is high are solved.

The invention discloses a fine velocity modeling method based on interval velocity inversion. The method comprises the following steps: carrying out collaborative modeling by utilizing an earthquake stacking velocity spectrum and a ground fitting velocity curve to obtain an interval velocity inversion initial model, determining a velocity anomalous body in an overlying stratum of a target stratum, carrying out tracking interpretation to obtain a structural interpretation layer, and taking the structural interpretation layer and the target stratum as control horizon in an inversion process, so as to obtain an interval velocity inversion model; and carrying out interval velocity inversion by utilizing depth information of a control horizon, a ground fitting velocity curve and an interval velocity inversion initial model to obtain interval velocities of a structural interpretation layer and a target layer, and meanwhile, fusing the interval velocities with an initial interval velocity of an overlying stratum of the structural interpretation layer to obtain a complete stratum interval velocity. The method not only strictly monitors the input data, but also considers the influence of the velocity anomalous body on interval velocity inversion, the interval velocity obtained by the method better conforms to geological conditions, the obtained structural map is higher in precision, and the success rate of well drilling development is improved.

The application provides a silt geology building pile foundation positioning method, and belongs to the technical field of pile foundation construction. The silt geology building pile foundation positioning method comprises the following steps: determining a main target detection area and a subordinate target area of silt geology; controlling a seismic source to send seismic waves from the subordinate target area to the main area; obtaining a first first-arrival time of the seismic waves from any geophone; determining a first target velocity of seismic wave propagation at each position in the main target detection area and a velocity inversion graph by using a seismic wave tomography method according to the first first-arrival time; determining the range of a silt geology high stress area in the velocity inversion graph according to the first target velocity and the corresponding relationship between velocity and stress; comparing the values of the vibration sensors in each detection hole in the high stress area range to determine the best piling position. The application solves the problems that the existing silt geology building pile foundation positioning method cannot quickly determine the range of the high stress area and cannot determine the best piling position.

The application provides a reflection structure constrained seismic data low frequency recovery method and device, belonging to the field of seismic data processing, comprising the following steps: determining the to-be-recovered highest frequency corresponding to the to-be-recovered low frequency seismic data according to original seismic data; determining a reflection structure representation operator by using the original seismic data and an inversion algorithm; constructing a regularization constraint term by using the reflection coefficient and the reflection structure representation operator, and establishing a target functional of the reflection coefficient; inversely solving the target functional of the reflection coefficient to obtain the reflection coefficient; performing low frequency filtering on the reflection coefficient to obtain low frequency seismic data; performing high pass filtering on the original seismic data to obtain high frequency seismic data; and combining the low frequency seismic data and the high frequency seismic data to obtain seismic data after low frequency recovery. Through the method provided by the application, the reliability of the low frequency seismic data can be ensured, the seismic data resolution is enhanced, and the accuracy of full waveform velocity inversion is improved.

The present invention relates to a method for accurately delineating the sand-mud transition zone of a uranium reservoir based on three-dimensional seismic velocity inversion, comprising the steps of positioning and calibrating the uranium reservoir based on well seismic data, implementing structural interpretation of the target layer and constructing an initial inversion model, implementing high-resolution velocity inversion to obtain a velocity body and quality control, filtering the velocity body to obtain a low-frequency velocity body, maximizing the dominant velocity value of the ore-bearing sand-mud transition zone, obtaining a body characterizing the sand-mud transition zone attribute, and delineating the sand-mud transition zone. This method achieves a three-dimensional depiction of the sand-mud transition zone in full three-dimensional space; it can analyze the spatial configuration relationship between the ore body and the sand-mud transition zone, thereby facilitating the summary of the corresponding mineralization laws and mineralization-controlling factors; the sand-mud transition zone finally delineated is of great significance for the comprehensive analysis of the mineralization mechanism and distribution law of sandstone-type uranium deposits in the sand-mud transition zone, and provides strong technical support for enriching the overall mineralization theory of sandstone-type uranium deposits and establishing prospecting models.

The application discloses a method for pre-stack seismic velocity inversion of natural gas hydrate reservoir based on deep learning. According to the occurrence characteristics of natural gas hydrate and free gas under actual geological conditions, combined with the real seabed layered sedimentary area and the public ocean velocity model, a synthetic velocity model of thin layer associated structure containing high and low speed alternating hydrate and free gas is constructed; the model is forward by solving the wave equation to generate forward seismic records; the forward seismic records and the synthetic velocity model are registered, and the training, testing and verification sets are divided; the deep neural network is constructed, and the training environment and hyperparameters are configured; the perception loss is introduced, the mixed loss function is designed, the network training is multi-dimensionally constrained, the network is optimized and the optimal parameters are saved; the optimal parameters are called to batch inversion of seismic records, and high-precision velocity inversion and thin layer identification are realized. Through the above method, the inversion efficiency and reservoir characterization ability are improved, and effective technical means are provided for natural gas hydrate exploration.

The invention discloses an apparent shear wave velocity inversion method and system, a medium and a product, and relates to the field of seismic data processing, and the method comprises the steps: obtaining frequency dispersion curve data, and carrying out the preprocessing and sorting of the frequency dispersion curve data, and obtaining a frequency dispersion curve data sequence; calculating the sensitive depth corresponding to each frequency point in the frequency dispersion curve data sequence according to an empirical formula; carrying out abnormal point detection and elimination and differential inversion on the frequency dispersion curve data sequence to obtain an apparent transverse wave speed data sequence; smooth processing and boundary correction compensation are carried out on the transverse wave velocity data sequence; and according to the apparent shear wave velocity data sequence after the boundary correction compensation and the sensitive depth corresponding to each frequency point, obtaining a profile map of the apparent shear wave velocity changing along with the depth. According to the method, the obtained profile map with the apparent shear wave speed changing along with the depth is obviously smoother in a high-frequency area, and interlayer speed transition better conforms to geological continuity.

The application discloses a VSP acoustic wave full waveform inversion method of a double-branch physical driving cyclic neural network, applied to the field of seismic data processing, and aims at the problems of weak explanation and high calculation cost of the existing seismic velocity inversion technology in a full data driving mode. Since a wave equation describes the change of a wave field in time and space, a time stepping method is adopted in numerical simulation, that is, the wave field state of the next moment is calculated according to the state of the previous moment. Thus, the wave field state of the previous moment is taken as a hidden layer of the RNN, the acoustic wave equation is solved by finite difference, and the velocity parameter in the equation is taken as a trainable parameter in the physical driving RNN forward network. The shot record output by the physical driving RNN forward network each time is taken as a predicted value. The shot record output by the physical driving RNN forward network by setting a real velocity parameter is taken as an observed value. The process of correcting the velocity parameter by reverse propagation of the loss between the predicted value and the observed value is called a velocity inversion process.

The invention relates to the field of geophysical research, in particular to a full waveform inversion method and device based on a time difference objective function. The method comprises the following steps: determining a time difference objective function according to an error between observed seismic data and simulated seismic data; determining an adjoint source corresponding to the time difference target function according to the target function, a first sampling time interval of the observation seismic data and a second sampling time interval of the simulation seismic data; taking the adjoint source as a seismic source to excite to form an adjoint wave field, and obtaining a gradient of full waveform inversion based on cross-correlation of the adjoint wave field and a seismic source wave field; and according to the gradient of the time difference target function, updating the iteration initial speed until an iteration condition is met, and obtaining a target speed model. According to the time difference objective function and the full-waveform inversion method, the convergence interval of a traditional objective function is widened, the cyclic wave jump suppression capability of full-waveform inversion is enhanced, and the speed inversion precision is improved.

The application discloses a deep-water shallow seismic data structure-constrained attenuation compensation velocity modeling method, which comprises the following steps: obtaining an initial velocity model by using a tomographic inversion method and performing reverse-time migration imaging to obtain an initial imaging result; and obtaining a seismic data dip angle field by using a dip angle prediction technology; a structure-constrained velocity modeling target functional is established by using the dip angle field and multi-frequency band seismic data; wave field continuation for Q compensation is performed by using a decoupled viscous wave equation, and the gradient of the functional is calculated to update the velocity model; a multi-scale multi-frequency band inversion strategy is used to solve a high-precision velocity inversion optimization problem to obtain a high-precision deep-water shallow velocity field. By applying attenuation compensation to full waveform inversion, a structure guide constraint is introduced into a full waveform inversion target function, and a velocity parameter model is constrained by using seismic imaging structure information, so that the precision and efficiency of inversion can be effectively improved.

The invention provides a Rayleigh surface wave shallow surface velocity inversion method and device which are applied to the technical field of engineering geophysical prospecting, and the method comprises the steps: building a preset number of shallow surface velocity models; performing forward modeling on the shallow surface velocity model through a finite difference method to obtain time domain waveform data of different receiving points; performing cross-correlation processing and frequency spectrum calculation on the time domain waveform data of different receiving points to obtain a frequency spectrum of a cross-correlation gather; performing frequency-Bessel transformation on the frequency spectrum of the cross-correlation gather to obtain a Rayleigh surface wave frequency spectrum; training a preset cyclic adversarial neural network through a preset one-dimensional speed structure and the Rayleigh surface wave spectrum to obtain a trained cyclic adversarial neural network; a first generator of a first generative adversarial network of the trained cyclic adversarial neural network is used as a prediction network, and the prediction network is used for carrying out speed inversion on an actual Rayleigh surface wave frequency spectrum; according to the invention, automatic inversion of the surface wave spectrum can be realized.

The invention belongs to the technical field of multiple wave imaging based on a one-way acoustic wave equation, and relates to a multiple wave joint migration inversion method aiming at a stratum attenuation effect, aiming at the attenuation effect existing in actual seismic exploration and combining the multiple wave imaging advantages of a joint migration inversion algorithm, the deep part energy and the resolution of a reflectivity migration ground are improved, and the stratum attenuation effect is improved. The shallow artifacts of the speed inversion result are eliminated, and the precision of the speed model is improved. The method comprises the following steps: picking up an initial velocity model and a reflectivity model through preprocessing according to obtained observation data; in each iteration, simulation data are obtained through full wave field simulation containing an attenuation effect, a wave field residual error is obtained by making a difference between the simulation data and observation data, and linear iteration updating is carried out on parameters through a gradient descent method; the reverse extrapolation of the wave field residual error can be mapped on the gradient of the propagation operator, and the gradient is converted into a velocity gradient through linear parameterization; in continuous iteration, the residual error between simulation data and observation data is minimized, and a satisfactory imaging result is obtained.

The present application relates to a kind of microseismic velocity inversion methods of deep learning and wave equation combined drive, underground model is dispersed, and setting detector;Establish underground velocity model, utilize finite difference algorithm to calculate the waveform information of random seismic source at each detector position;Waveform record, velocity model and seismic source coordinate are batched into Unet neural network;Pure data driven calculation obtains preliminary velocity structure, physical drive calculation, uses finite difference to solve wave equation, obtains seismic record;Loss function is composed of two parts of velocity error and waveform error;The structural similarity between the predicted velocity and the real velocity is calculated, and the average value is set as the dynamic weight of physical drive loss value;The mapping relationship between waveform feature and velocity model is extracted using neural network, and the underground velocity structure is inverted to predict.The present application realizes artificial intelligence real-time prediction microseismic velocity model, and the precision of inversion velocity model is high, and the interpretability is strong.

The application relates to a method for improving the imaging precision of deep reflection seismic data. The method comprises the following steps: performing first pre-stack depth migration on an initial depth domain layer velocity model and a common center point gather to obtain a first pre-stack depth migration profile and a pre-stack depth migration gather; picking a seed point by using the first pre-stack depth migration profile, picking residual curvature by using the pre-stack depth migration gather, performing velocity inversion on the initial depth domain layer velocity model according to the seed point and the residual curvature to obtain an optimized layer velocity model; performing second pre-stack depth migration on the optimized layer velocity model to obtain a second pre-stack depth migration profile; and if the migrated layer velocity converges, the second pre-stack depth migration profile is the imaging of the deep reflection seismic data. The scheme provided by the application can better improve the imaging precision of the deep rock structure and reveal the deep tectonic characteristics of the lithosphere.

The present application relates to a kind of seismic velocity inversion method, belong to oil and gas geophysical engineering field.By the amplitude maximum of each shot estimated from the initial observed seismic data shot point position obtained, the combined time shift parameter of iteration process is determined by the maximum time shift parameter of combined stack and total iteration number, the combined single shot gather of iteration is generated by using combined time shift parameter to single shot delay stack, the combined seismic source is generated by using combined time shift parameter to seismic wavelet delay stack, the new simulated single shot is obtained by using combined seismic source initial velocity model and observation system forward, the proportion coefficient is determined by the amplitude of new simulated single shot and the amplitude of initial observation data, the amplitude of combined seismic source after forward is updated by proportion coefficient, the forward combined single shot is obtained by updating combined seismic source, initial velocity field and again forward, the final velocity field is obtained by the cross-correlation gradient of forward combined single shot and combined single shot gather, the energy difference of each shot is fully considered, and the stability and accuracy of inversion are improved.

An active and passive source combined geological detection method for a shield tunnel. The method comprises: acquiring an active-source seismic-wave reflected signal and a passive-source acoustic-wave reflected signal; on the basis of the active-source seismic-wave reflected signal, obtaining an active-source geological detection imaging result by using a longitudinal and transverse wave combined imaging principle; on the basis of the passive-source acoustic-wave reflected signal, obtaining a passive-source geological detection imaging result by using a frequency dispersion extraction-transverse wave velocity inversion method that fuses spatial autocorrelation and a half-wavelength model; and by using the active-source geological detection imaging result as a main part and the passive-source geological detection imaging result as an auxiliary part, and in view of an adaptive adjustment coefficient therebetween, obtaining an adverse geological risk level of a region to be detected.

The present application belongs to the technical field of geological disaster monitoring, and discloses a debris flow velocity inversion method based on an optical flow model, and proposes an integrated optical flow framework for simultaneously inverting the surface velocity field and the front velocity of the debris flow. Through three sets of large-scale flume experiments, the method provides a robust and accurate tool for characterizing complex flow kinematics. By simultaneously measuring the surface velocity field and the front velocity, the main limitations of existing computer vision techniques are overcome, and a more comprehensive understanding of the flow process is achieved. The core scientific contribution lies in the new physical insights obtained through synchronous measurement: the research results provide direct and strong observational evidence, which links the high-frequency velocity pulses at the front to the secondary surge dynamics propagating from the tail. This discovery is crucial for understanding internal momentum transfer and energy dissipation. At the methodological level, the framework is superior to the benchmark deep learning tracker in terms of reproducing flow velocity statistical characteristics and ensuring numerical accuracy.

The invention relates to the field of ocean geophysical exploration, and provides a near-seabed transverse wave exploration method and system based on an active source earthquake, and the method comprises the steps: collecting seismic data based on an active source earthquake system; interface waves recorded by the vertical component detector and the hydrophone component detector are separated; frequency dispersion curve energy spectrums of the interface waves recorded by the vertical component detector and the interface waves recorded by the hydrophone component detector are calculated respectively; obtaining an optimal weighting coefficient; obtaining a superposed frequency dispersion curve energy spectrum; and based on a half-wavelength method of Poisson's ratio constraint, carrying out inversion by using fundamental frequency information in the superposed frequency dispersion curve energy spectrum, and obtaining a transverse wave velocity structure of a near-seabed area below the measuring line. According to the method, the frequency dispersion curve is extracted by preferably selecting the data of the vertical component geophone and the data of the hydrophone component geophone with prominent fundamental frequency energy, and the transverse wave velocity inversion is carried out, so that a high-quality data basis is provided; the Poisson's ratio is introduced into inversion, and system errors caused by model simplification are effectively reduced.

The application discloses a kind of inversion initial model fast establishment, inversion method and system of seismic transverse wave velocity, comprising: first, extracting surface wave dispersion curve from original surface wave data, and the phase velocity in surface wave dispersion curve, frequency is converted into apparent transverse wave velocity and depth.Then, based on the converted surface wave dispersion curve, according to the depth corresponding to the layer depth array in initial model, the apparent transverse wave velocity in the converted surface wave dispersion curve is assigned to each layer depth array using interpolation algorithm, and the inversion initial model of seismic transverse wave velocity is established.Because the apparent transverse wave velocity and the actual transverse wave velocity are consistent in variation trend, the inversion initial model thus established can ensure the convergence of algorithm to remain stable when seismic transverse wave velocity is inverted, and avoid inversion into local minimum value or not convergent.

The invention provides a drilling geological engineering risk early warning method based on well-seismic integrated parameters, and the method comprises the steps: obtaining the abnormal main control factors of an easy-to-collapse and easy-to-leak section based on the drilling characteristics, logging characteristics and logging characteristics of the easy-to-collapse and easy-to-leak section; performing full-strata lithology prediction on the to-be-drilled well to obtain the distribution conditions of mudstone, sandstone and coal seam; according to the distribution conditions of the mudstone, the sandstone and the coal seam, abnormal main control factors corresponding to the lithology are predicted; performing full-strata velocity inversion on the to-be-drilled well, and predicting an elastic parameter; according to the prediction condition of the elastic parameters, predicting the formation pressure, collapse pressure, leakage pressure and fracture pressure of the to-be-drilled well; and constructing a risk prediction model by integrating the prediction conditions of the pore pressure, the collapse pressure, the leakage pressure and the fracture pressure of the full strata of the to-be-drilled well and the prediction conditions of the abnormal main control factors corresponding to the lithology. The method achieves the timely recognition of exploration risks, improves the operation efficiency, and guarantees the operation safety.