32 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 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 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 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 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.

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 application discloses a seismic wave velocity inversion method based on function smooth noise suppression technology, comprising the following steps: defining seismic wave motion equation parameters and boundary conditions; generating a training data set; constructing a full connection neural network; defining a total loss function L; training the full connection neural network to convergence by using the training data set to minimize L, and obtaining a seismic wave propagation velocity solving model; and solving the predicted seismic wave propagation velocity of any space-time point in a research area by using the seismic wave propagation velocity solving model. The method can fully utilize observation data to quickly and accurately calculate the predicted seismic wave propagation velocity of any space-time point in the research area, can dynamically adapt to the performance of the space-time point in training, does not need to preset noise prior knowledge, avoids subjective estimation of noise level, effectively reduces the influence of noise, can obtain results at a lower calculation cost even in a high-dimensional case, and reduces the overfitting risk while retaining physical consistency.

The invention relates to a seismic velocity inversion method, and belongs to the technical field of seismic exploration. According to the seismic velocity inversion method, seismic data are divided into different frequency bands, and then full-waveform inversion is carried out by using a low-frequency band to obtain a velocity field; fusing the velocity field with the well velocity field, and performing grid chromatography inversion to obtain a well constraint velocity field; and finally, carrying out grid chromatography inversion and full waveform inversion on the well constraint velocity field to obtain a target velocity field. The seismic velocity inversion method can solve the problem of cycle jump caused by low-frequency loss, has good adaptability to data with low-frequency loss in seismic data, and can improve the stability and accuracy of the seismic data inversion velocity field.

The invention relates to the technical field of seismic data processing, and particularly discloses a fault velocity inversion method and device based on amplitude coherence attribute localization, and the method comprises the steps: obtaining a fault development position through the amplitude coherence attribute of a migration imaging data body; the range distance influenced by the fault velocity model is judged based on the fault development position and the migration imaging profile; picking up the residual time difference of the common imaging point gather based on the range distance; and performing grid chromatography velocity iteration based on the residual time difference to obtain an optimized fault velocity model. The fault development position is identified through the amplitude coherence attribute, the speed range influenced by the fault is comprehensively judged according to the fault distribution characteristics and the common reflection point depth gather, the common reflection point remaining gather in the effective range is used for grid tomography inversion, the update quantity of the speed model near the fault can be obtained, and the fault development speed is improved. The method improves the speed model precision of a complex structure region with fault development, and is more beneficial to correct homing of the fault and accurate implementation of the structure form near the fault.

The application discloses a VSP hourly velocity inversion method of a physical embedding recurrent neural network, applied to the field of seismic velocity inversion, and aims at the problem of low precision of existing seismic velocity inversion methods; on the basis of realizing the embedding of a recurrent neural network of a finite difference acoustic wave equation, the application modifies the back propagation algorithm of the RNN, calculates the residual error of a current wave field and a shot gather wave field at each independent time, takes the residual error as the loss at the current time, and sums the loss at the current time and the loss at the previous time at each time, so as to update the velocity model in the residual error updating range until the residual error meets the condition, and then the velocity inversion area is expanded, so that the step-by-step correction from the shallow part to the deep part is realized; compared with other inversion correction algorithms, the hourly inversion technology proposed by the application significantly improves the accuracy and reliability of the physical embedding RNN velocity inversion.

The invention discloses a VSP pre-drilling formation velocity prediction and driving velocity modeling method. In order to solve the problems that in the prior art, an underground structure of a pre-drilling area is fuzzy, and the precision of a speed model is insufficient, the method comprises the steps that firstly, the acoustic logging speed is corrected through the high-precision longitudinal wave speed obtained through a zero well spacing VSP; secondly, according to the rock physical relationship between the speed and density of a target layer before drilling of an adjacent region and the first arrival travel time prediction before drilling, time-depth constraint is constructed in combination with VSP multi-wave information, and a pre-drilling region initial speed model is established; based on a VSP corridor and the corrected acoustic curve, obtaining an optimal seismic wavelet; and establishing a target function suitable for VSP pre-drilling speed inversion, and obtaining a pre-drilling speed inversion result by using VSP corridor data. And finally, under the constraint of a stratigraphic structure obtained by ground seismic data interpretation, finishing updating of a ground seismic velocity model by using the velocity obtained by VSP zero offset and pre-drilling inversion, so that the velocity model is more matched with underground real velocity structure characteristics.

The invention discloses a natural gas hydrate reservoir pre-stack seismic velocity inversion method based on deep learning. The method comprises the following steps: according to occurrence characteristics of natural gas hydrate and free gas under actual geological conditions, combining a real seabed layered deposition area and a public ocean speed model, and constructing a synthetic speed model of a thin-layer associated structure containing high-speed and low-speed alternating hydrate and free gas; performing forward modeling on the model by solving a wave equation to generate a forward modeling seismic record; registering a forward modeling seismic record and a synthetic velocity model, and dividing training, testing and verification sets; constructing a deep neural network, and configuring a training environment and hyper-parameters; introducing perception loss, designing a mixed loss function, carrying out multi-dimensional constraint network training, optimizing the network and storing optimal parameters; and calling the optimal parameters to perform batch inversion on the seismic records to realize high-precision speed inversion and thin-layer identification. According to the method, the inversion efficiency and the reservoir characterization capability are improved, and an effective technical means is provided for natural gas hydrate exploration.

The invention provides an early-arrival wave equation travel time inversion method for optimizing an error function. The early-arrival wave equation travel time inversion method for optimizing the error function comprises the following steps: step 1, inputting field observation early-arrival wave seismic data, seismic wavelets and an initial speed model; step 2, carrying out numerical solution on the wave equation to obtain a seismic source wave field and simulated seismic data; 3, calculating the travel time difference between the observed seismic data and the simulated seismic data, and constructing an accompanying seismic source; 4, the wave equation is subjected to numerical solution, and an accompanying wave field is obtained; step 5, calculating a gradient and obtaining a speed update quantity; step 6, constructing an optimized error function according to the travel time difference and the seismic data; and step 7, when the inversion result converges, outputting the inversion result. According to the early arrival wave equation travel time inversion method for optimizing the error function, the defect that wave equation inversion is prone to falling into a local minimum value is overcome, and inversion convergence and speed inversion precision are improved.