Inverse Method Research Articles

Simultaneous quantification of triple pore types in carbonate reservoirs using well logs and seismic data

Accurate characterization of carbonate pore types is of great significance for seismic reservoir characterization. However, the existing pore-type inversion methods cannot characterize the coexistence of three main pore types (vuggy pores, matrix pores, and cracks) for each modeling point because they apply a simple iterative procedure without using S-wave velocity. To overcome this problem, we develop a simultaneous triple pore-type inversion (TPTI) strategy to estimate the porosities of different pore types from the P- and S-wave velocities using a global optimization technique known as very fast simulated annealing. Considering the rock composition variation and pore geometry complexity, we first establish an easy-to-implement multipore effective medium model for carbonate reservoirs based on the Keys-Xu model and Gassmann’s equation. Then, the applicability of this model is verified by two sets of experimental acoustic measurements for synthetic and actual carbonate samples. A good comparison of model predictions with laboratory data indicates that our model can effectively capture the elastic responses of carbonate rocks with different pore shapes, which provides a theoretical basis for quantifying multiple pore types. Based on the developed model, we further develop a TPTI method. We apply it to describe pore-type distribution from well logs and seismic data in a heterogeneous carbonate reservoir from the Sichuan Basin in southwest China. Real applications suggest that our method significantly improves the accuracy of porosity estimates for different pore types, outperforming the traditional dual pore-type inversion method. This advancement holds considerable promise for the seismic characterization of pore types in ultradeep carbonate reservoirs.

Damage Evolution Mechanism of Railway Wagon Bogie Adapter 1035 Steel and Damage Parameter Calibration Based on Gursone-Tvergaarde-Needleman Model.

As a critical component of a train, the railway wagon bogie adapter has higher quality requirements. During the forging process, external loads can induce voids, cracks, and other defects in the forging, thereby reducing its service life. Hence, studying the damage behavior of the forging material, specifically AISI 1035 steel, becomes crucial. This study involved obtaining stress-strain curves for AISI 1035 steel through uniaxial tensile tests at temperatures of 900 °C, 1000 °C, and 1100 °C, with strain rates of 0.1 s-1, 1 s-1, and 10 s-1. Subsequently, SEM was used to observe samples at various deformation stages. The damage parameters, q1, q2 and q3 in the GTN model "a computational model used to analyze and simulate material damage which can effectively capture the damage behavior of materials under different loading conditions" were then calibrated using the Ramberg-Osgood model and stress-strain curve fitting. Image Pro Plus software v11.1 quantified the sample porosity as f0, fn, fc and fF. A finite element model was established to simulate the tensile behavior of the AISI 1035 steel samples. By comparing the damage parameters of f0, fn, fc and fF obtained by the finite element method and experimental method, the validity of the damage parameters obtained by the finite element inverse method could be verified.

Characterization and Inversion Method of Relative Permeability of Emulsion Flooding.

Relative permeability curves are one of the crucial factors that control the behavior of multiphase flow in petroleum reservoirs. Accurate prediction of the relative permeability curve involving emulsification and emulsion transport in emulsion flooding is vital to enhancing heavy oil recovery. In this study, a relative permeability characterization model including the emulsification mechanism and emulsion transport is developed, in which residual oil saturation and end-point permeability for water are parametrized as functions of emulsion concentration and capillary number. Based on the built characterization model, an inversion method (ensemble Kalman filter) is used to estimate the relative permeability curves of emulsion flooding. The inversion results of four sets of emulsion flooding experiments show that the errors of end points between the Johnson-Bossler-Neumann method and inversion method are less than 10%. For the continuous emulsification process in the in situ emulsion flooding tests, the inversed relative permeability curves shift between the upper and the inferior limits of the permeability. The error of residual oil saturation between the inversion and the experiment is less than 10%. In summary, this study demonstrates the feasibility of using the characterization model and inversion method to infer the relative permeability curves of emulsion flooding, which can aid in the predictions of multiphase flow in heavy oil reservoirs.

Noise suppression of infrared thermal imaging of rocket exhaust plume using SPOD

The environmental noise have a negative influence on the quality of infrared thermal imaging of the rocket exhaust plume. In this study, the noise data of the unsteady rocket exhaust plume flow field was simulated using Gaussian white noise, and the infrared thermal image of the plume was numerically calculated using the narrow band method (SNB) and the line of sight (LOS) method. The denoising of infrared thermal imaging was achieved through the spectral proper orthogonal decomposition (SPOD) inversion method. Results indicate that Gaussian white noise leads to larger infrared thermal image residuals in the intrinsic core of the plume compared to the mixed regions. The infrared thermal image in the 2.7 μm band is greatly affected by the noise with an average error of 21.1%, and the average error in the 4.3 μm band is 17.6%. After SPOD denoising, the error of the plume infrared thermal image is reduced by more than 50%.

Comparison of terrestrial exospheric hydrogen 3D distributions at solar minimum and maximum using TWINS Lyman-α observations

Remote sensing observations of far-ultraviolet (FUV) emissions have been used to estimate 3D neutral hydrogen (H) density models of the terrestrial exosphere under solar minimum (2008) and solar maximum (2013 and 2015) conditions. Specifically, we used Lyman-alpha (Lyman-α, FUV at 121.6 nm) radiance data acquired by the Lyman-alpha detectors (LADs) onboard NASA’s TWINS satellites, collected over several days for each of the 3 years to provide sufficient coverage of the exospheric region. The datasets included Lyman-α measurements taken only above Earth radii (Re) of 3.75, assumed to be the optically thin region of the exosphere, where the measured Lyman-α intensity along a line of sight (LOS) is linearly proportional to the atomic hydrogen column density. Based on the calibration using multiple UV-bright stars, a significant decrease in TWINS1 LAD1/2 sensitivities was found for 2013 (LAD2 ∼1/2, LAD ∼1/3) and 2015 (LAD2 ∼1/3, LAD ∼1/7) compared to 2008. The calibration uncertainty was derived to be, on average, 5%. We estimated the 3D global hydrogen density distributions from these radiance data using two different tomographic inversion methods: first, a parametric fitting method based on a spherical harmonic function of order 3 and second, a high degree-of-freedom retrieval approach to validate our retrievals of H density. Both inversion methods consistently incorporate the effects of Lyman-α absorption within the exosphere and Lyman-α re-emission from Earth’s albedo. Our results reveal that H densities during solar maximum conditions are, on average, 30%–40% higher than those during solar minimum. All models showed a high concentration of atomic H on the dayside and nightside near the Sun–Earth line, which determines a nose/geotail structure consistent with theoretical effects from the solar radiation pressure. Furthermore, we identified that H-density enhancements during solar maximum with respect to solar minimum conditions occur at mid-to-high latitudes, particularly on the dusk side, while no significant enhancement seems to occur near the dayside nose.

Efficiency improvement and pressure pulsation reduction of volute centrifugal pump through diffuser design optimization

To investigate the effects of load distribution parameters of the diffuser vanes on the efficiency and pressure pulsation of the volute centrifugal pump, the diffuser shape was designed using an inverse design method, and an optimization method was proposed for the diffuser blade load distribution based on the combination of Latin hypercubic sampling, artificial neural network, and non-dominated sorting genetic algorithm-II. Steady and unsteady simulations were performed by solving the three-dimensional Reynolds-averaged Navier–Stokes equations using the shear stress transport k−ω turbulence model. The results obtained from Pearson correlation analysis revealed that the slope values Ks and Kh of the shroud and hub load distribution curves had the most significant effect on the pump head and efficiency. Compared with those of the original model, the efficiency and head of the optimized volute centrifugal pump under the design conditions increased by 3.8 and 3.5 %, respectively. The pressure pulsation intensities in the vaneless regions were effectively mitigated for the optimized volute centrifugal pump. The pressure pulsation coefficients at three times the diffuser vane frequency (3fDIF) reduced by 58.5 and 43.8 % at monitoring points M1 and M2, in the impeller outlet, respectively. The pressure pulsation coefficients at the impeller blade passing frequency (fBPF) reduced by 6.5 and 14.4 % at monitoring points M3 and M4 in the diffuser inlet, respectively. The results showed that the optimized diffuser improved the overall performance of the volute centrifugal pump, and demonstrated the feasibility of the optimization method.

Adaptive resolution force probe simulations: Coarse graining in the ideal gas approximation.

The unfolding of molecular complexes or biomolecules under the influence of external mechanical forces can routinely be simulated with atomistic resolution. To obtain a match of the characteristic time scales with those of experimental force spectroscopy, often coarse graining procedures are employed. Here, building on a previous study, we apply the adaptive resolution scheme (AdResS) to force probe molecular dynamics (FPMD) simulations using two model systems as examples: One system is the previously investigated calix[4]arene dimer that shows reversible one-step unfolding, and the other example is provided by a small peptide, a β-alanine octamer in methanol solvent. The mechanical unfolding of this peptide proceeds via a metastable intermediate and, therefore, represents a first step toward a complex unfolding pathway. We show that the average number of native contacts serves as a robust order parameter for the forced unfolding of this small peptide. In addition to increasing the complexity of the relevant conformational changes, we study the impact of the methodology used for coarse graining. Apart from the iterative Boltzmann inversion method, we apply an ideal gas approximation, and therefore, we replace the solvent by a non-interacting system of spherical particles. In all cases, we find excellent agreement between the results of FPMD simulations performed fully atomistically and those of the AdResS simulations also in the case of fast pulling. This holds for all details of the unfolding pathways, such as the distributions of the characteristic forces and also the sequence of hydrogen-bond opening in case of the β-alanine octamer. Therefore, the methodology is very well suited to simulate the mechanical unfolding of systems of experimental relevance.

Waterborne novolac epoxy‐based thermal resistant and fire‐retardant thermal insulation coatings

AbstractWaterborne novolac epoxy based thermal resistant and fire‐retardant thermal insulation coatings were developed in this work. Novolac (phenolic) epoxy emulsion (PEE) was prepared by the phase inversion method based on formulation optimization by orthogonal array testing. The prepared emulsions have similar particle size, viscosity, and stability to commercial bisphenol a glycidyl ether epoxy emulsion (PZ). DSC results showed that the varnish film based on PEE had a higher glass transition temperature and initial thermal decomposition temperature compared with PZ cured by the same curing agent. Waterborne thermal insulation coating with PEE and PZ as binders and hollow glass microspheres (HGMs) as thermal insulation fillers were studied. The results indicated that the thermal conductivity of the coating decreased with an increase in HGMs content. Specifically, the thermal conductivity of the PEE coating containing 20 wt% HGMs is as low as 0.093 W/(m·K). It reduces the exterior temperature of the hot tank from approximately 200 °C to approximately 106 °C. Furthermore, the HGMs/PEE waterborne thermal insulation coating exhibits good thermal stability and flame retardance.

Time-lapse difference seismic inversion based on scattering theory

Time-lapse seismic monitoring technology is an effective tool for quantitatively estimating subsurface medium variations and plays a crucial role in resource and engineering explorations. Among existing time-lapse inversion methods, the difference inversion technique offers higher accuracy but requires linear inversion operators. During linearization, the background is often approximated as a homogeneous medium, resulting in the loss of effective information. Additionally, merely using reflection wave information is not conducive to improving inversion accuracy. However, the true structure of the subsurface is often complex and inhomogeneous, which can easily generate complex wavefields, such as scattering waves, thereby hindering the resolution of differences between time-lapse models. Therefore, incorporating more comprehensive wavefield information is essential for enhancing the accuracy of time-lapse seismic inversion. To address this issue, a wave-equation-based time-lapse difference inversion method is proposed to retrieve the production-induced property perturbation. In the proposed method, the forward operator is formulated through the scattering wave equation, the operator is linearized through the Born approximation, and the WKBJ method is used to approximate Green's function with high accuracy for the inhomogeneous background. In this study, time-lapse seismic parallel inversion and difference inversion are conducted using the proposed forward operator. Both numerical experiments and application of field seismic data indicate that the proposed time-lapse difference inversion method can realize a more accurate quantitative characterization of time-lapse variations.

Geostatistical Inversion of Frequency-Domain Electromagnetic Data for Near Surface Modelling

The detailed characterization of near-surface deposits is important for both environmental and economic reasons. These shallow subsurface systems can be very complex and heterogenous due to natural dynamics and anthropogenic interferences. Modeling techniques based exclusively on direct sampling generate limited informed three-dimensional models of the near-surface. Geophysical methods provide valuable and additional information to model the spatial distribution of the near-surface for locations where direct observations are not available. From this set of methods, frequency-domain electromagnetic induction (FDEM) has been successfully applied to image complex near-surface deposits. Yet, predicting the spatial distribution of relevant subsurface properties from geophysical data, and the integration of direct observations, is not straightforward. It requires solving a challenging geophysical inversion problem. Geostatistical modelling tools have been effectively applied to couple direct observations with geophysical data such as seismic reflection. We propose an iterative geostatistical FDEM inversion method able to integrate data from direct measurements of the near-surface with surface loop-loop FDEM measurements to simultaneously predict high-resolution models of electrical conductivity and magnetic susceptibility, and their associated uncertainty. The iterative geostatistical inversion method is based on stochastic sequential simulation and co-simulation as model perturbation and update techniques. The iterative optimization is based on the local data misfit between observed and simulated FDEM data, weighted by the sensitivity of the acquisition equipment. The proposed method is first demonstrated for a synthetic landfill data set created based on real data collected at a mine tailing disposal site in Portugal, and on a real data set collected at a site with archaeological features in Knowlton, UK. The results show the ability of the proposed method to accurately predict and characterize the spatial distribution of electrical conductivity and magnetic susceptibility down to the depth of interest while reproducing the recorded FDEM data.

Changes of the state in the commodity economy

The purpose of this article is to clarify the transformation of the state in the commodity economy, including the transformation from class domination to human management, which is the transition process from the division of labor to the regulation and distribution of society by monetary policy. To achieve this purpose, the content is approached from the methods of humanistic philosophy and social philosophy, and at the same time, uses qualitative, analytical, synthetic, and comparative methods, especially the inversion method, to clarify the transformation between the subject and the product in the means and purposes of the existence and development of the state. The results affirm that the state is a product established with a specific historical nature to serve human life. However, when each individual is a product, it carries the instinct of that state, depending on the state. Conclusion: First, the state is a product created by humans, established and produced to serve human life. Second, the state's primary function is to enforce justice, but it becomes the job of civil servants. Third, the function of the state in a commodity economy is to manage and regulate the commodity economic process according to the law and to direct and distribute it by money. Fourth, human rights are the basis for the existence and transformation of the state in a commodity economy.

Prediction of shale bedding-parallel crack density based on the SH-SH wave inversion of the VTI media

Bedding-parallel cracks relate to hydrocarbon accumulation and can be beneficial for hydraulic fracturing. The prediction of bedding-parallel crack density (BCD) from seismic data is difficult because the development of bedding-parallel cracks has limited effects on the velocities and density of subsurface formation. Well logs of the Qingshankou (K2qn) shales in the Songliao Basin, northeast China, reveal that the target formations represent extremely strong seismic anisotropy (> 40%), and the magnitude of S-wave anisotropy highly relates to the density of bedding-parallel cracks (up to 2500 m−1). We present a novel BCD prediction method using the reflected seismic shear wave. The S-wave anisotropy parameter γ is first inverted from the SH-SH wave by using a stepwise inversion method based on a new shear wave reflection coefficient approximation. Furthermore, the highly correlated rock-physics relationship between BCD and the anisotropy parameter γ facilitates accurate BCD estimation. Field data application on a real seismic shear wave data set acquired from the study area demonstrates the viability of the proposed method. The investigation provides a useful tool for the characterization of ‘sweet-spots’ of shale-oil reservoir formation.

Design and performance study of a multiband metamaterial tunable thermal switching absorption device based on AlCuFe and VO2.

In this paper, we propose a multiband adjustable metamaterial absorption device based on a Dirac semimetal (BDS) AlCuFe and a thermally controlled phase-change material VO2. The absorption device has an axially symmetric structure, resulting in polarization-independent characteristics, and when VO2 is in a high-temperature metal state, ultra-high absorption rates and sensitives at frequencies of M1 = 2.89 THz, M2 = 7.53 THz, M3 = 7.97 THz, and M4 = 9.02 THz are achieved. Using a parameter inversion method, we calculated the impedance of the absorber, proving that it achieves impedance matching and produces perfect absorption in the resonance region. Additionally, we changed the physical and chemical parameters of the absorption device, demonstrating the device's excellent tunability and manufacturing tolerance. Furthermore, by lowering the temperature of VO2 to that of a low dielectric state, additional resonant peaks with ultra-high absorption rates at frequencies M5 = 5.62 THz, M6 = 7.16 THz, M7 = 7.64 THz, and M8 = 8.80 THz were obtained, broadening the absorption band of the device. Lastly, we investigated the detection sensitivity of the device by changing the external refractive index, resulting in a maximum sensitivity of 2229 GHz RIU-1. To sum up, the absorption device has great application potential in the fields of communication, sensing, temperature detection and photoelectric instruments.

Low-degree gravity field coefficients based on inverse GNSS method: insights into hydrological and ice mass change studies

The relative displacements of stations from a global network of Global Navigation Satellite System (GNSS) sites provide information on global mass transport. In this study, we use 19 years of global GNSS station displacements from the 3rd International GNSS Service reprocessing campaign to estimate the coefficients of the spherical harmonics of the Earth’s gravity field up to degree and order 8 using the inverse GNSS method based on elastic loading theory. The results indicate that the C30 coefficient can be derived based on GNSS station displacements as an alternative to solutions provided by Satellite Laser Ranging (SLR) and Gravity Recovery and Climate Experiment (GRACE). GNSS may support GRACE solutions that face the problems of deriving C30, which has fundamental meaning in estimating ice mass changes in polar regions. The recovery of Antarctic ice sheet mass change from January 2007 to December 2020 based on coefficients replaced by GNSS estimates results in a linear trend of − 152 ± 4 Gt/year, in comparison to − 149 ± 2 Gt/year for the replacement based on SLR from GRACE Technical Note #14. The results indicate that the spatial and seasonal patterns of terrestrial water storage changes derived from GNSS are consistent with those estimated using GRACE/GRACE Follow-On and SLR at a few-millimeter level in the Amazon and Brahmaputra River basin regions.

Distance-regularized level set inversion of magnetic data

Inversion is one of the key and difficult issues in potential field data processing and interpretation, and high-precision inversion has long been a popular research topic. We develop the distance-regularized level set inversion of the magnetic data (DRLSI-M) method. A double-well function is introduced into the objective function of the level set inversion, and we take advantage of its mathematical properties to minimize the gradient of the level set function so that it reaches a minimum value at points 0 and 1. As a result, the distance-regularized term not only replaces the costly reinitialization process so that the level set function maintains a signed distance property but also stabilizes the evolution of the level set function. When solving for the minimum value of the objective function, we transform the optimization problem into an initial-boundary value problem and solve it with the finite-difference method. The distance-regularized level set inversion method and the reinitialization level set inversion method are tested using three models: a single-level-set double-inclined plate model, a double-level-set double-inclined plate model, and a multiple-level-set three-cuboid model. Through model tests, the soundness of our method is verified and compared with the reinitialization level set inversion method. Our method avoids the limitations such as low efficiency and instability caused by the reinitialization of the level set inversion method. Furthermore, we apply our method for the inversion of aeromagnetic data from the Jinchuan copper-nickel mine. The results are consistent with the known geologic information, thus validating the practicality and effectiveness of DRLSI-M. The distance-regularized level set inversion method provides a theoretical basis for deep mine exploration.

A Novel Defect Quantification Method Utilizing Multi-Sensor Magnetic Flux Leakage Signal Fusion

In the assessment of pipeline integrity using magnetic flux leakage (MFL) detection, it is crucial to quantify defects accurately and efficiently using MFL signals. However, in complex detection environments, traditional defect inversion methods exhibit low quantification accuracy and efficiency due to the complexity of their algorithms or excessive reliance on a priori knowledge and expert experience. To address these issues, this study presents a novel defect quantification method based on multi-sensor signal fusion (MSSF). The method employs a multi-sensor probe to fuse the MFL signals under multiple lift-off values, enhancing the diversity of defect information. This enables defect-opening profile recognition using the characteristic approximation approach (CAA). Subsequently, the MSSF method is based on a 3D magnetic dipole model and integrates the structural features of multi-sensor probes to develop an algorithm. This algorithm iteratively determines the defect depth at multiple data acquisition points within the defect region to obtain the maximum defect depth. The feasibility of the MSSF quantification method is validated through finite element simulation and physical experiments. The results demonstrate that the proposed method achieves accurate defect quantification while enhancing efficiency, with the number of iterations for each defect depth calculation point consistently requiring fewer than 15 iterations. For rectangular metal loss, perforation, and conical defects, quantification errors are less than 10%, meeting practical inspection requirements.

A petrophysically driven seismic inversion method for the total organic carbon content of source rocks

Organic carbon content is the main index for evaluating organic matter abundance of source rocks, and it is still difficult to quantitatively predict total organic carbon (TOC) in source rocks based on seismic data. The study of seismic fluid identification driven by petrophysics can help to understand the fluid characteristics and distribution patterns of subsurface oil and gas reservoirs. First, this paper clarifies the sweet spot parameters (parameters characterizing hydrocarbon enrichment) and sensitive elastic parameters (a parameter characterizing the nature of an ideal elastic body) of source rocks through theoretical petrophysical modeling. Next, the paper establishes the relationship between sensitive elastic parameters and sweet spot parameters TOC and constructs a statistical petrophysical model that can characterize the relationship between the two on this basis. And then we construct the joint distribution of TOC and elastic impedance through the Bayesian theoretical framework to obtain the maximum posterior probability estimate as the final TOC inversion results of source rocks. Our method successfully predicts the spreading of high-quality source rocks in the Wen4 Section of Lufeng 13 Subsag, and the inversion results are within an uncertainty range of ±14 m for well data, which proves the reliability of the method. The prediction results show that the organic matter abundance of source rocks in the Wen4 Section is high, and the organic carbon content is generally higher than 2%, which provides a reliable basis for the further implementation of the resource scale of the depression and the clarification of the hydrocarbon-rich area, which provides technical support for the evaluation of the source rocks of the new depression in the new area.

2D Transdimensional Joint Inversion of Radio Magnetotelluric and Electrical Resistivity Tomography Data

Summary The joint inversion of radio magnetotelluric and electrical resistivity tomography data has the potential to reduce the uncertainties in the subsurface conductivity model. This is particularly beneficial when the datasets offer complementary information about the subsurface. However, the traditional gradient-based inversion methods pose challenges in quantifying uncertainty, as they yield a single model with limited appraisal of parameter uncertainty. The Bayesian inversion approach stands out for its capacity to provide quantitative assessments of uncertainty in the inverted model parameters. This is accomplished by generating an ensemble of models, leading to a posterior distribution that encapsulates both prior information concerning model parameters and the dataset information. We have implemented a transdimensional Markov chain Monte Carlo algorithm to perform the joint inversion of radio magnetotelluric and electrical resistivity tomography data. Through synthetic data studies, we illustrate how the inclusion of two complementary datasets can effectively reduce uncertainties in model parameters and how the model parameter uncertainties can be quantified. Subsequently, the developed algorithm is tested using exemplary field data from a waste site near Roorkee, India. Intensive prior geoelectric and transient electromagnetic as well as radio magnetotelluric studies investigated possible waste water seepage with a potential to contaminate the shallow aquifers. The derived subsurface structure from our transdimensional Bayesian results compare well with the deterministic results for the exemplary profile, but in addition provide comprehensive uncertainty estimates.

Pre-stack seismic inversion based on one-dimensional GRU combined with two-dimensional improved ASPP

Abstract Pre-stack seismic inversion is essential to detailed stratigraphic interpretation of seismic data. Recently, various deep learning methods have been introduced into pre-stack inversion, effectively capturing the vertical correlations of seismic data. However, existing deep learning methods face challenges such as insufficient feature extraction, poor lateral continuity, and unclear inversion details. We introduce the atrous spatial pyramid pooling (ASPP) module into the pre-stack inversion process, modifying the connection order and mode of its three components. Additionally, we incorporate a triplet attention module to extract features at different scales and utilize a gate recurrent unit (GRU) module to extract global information. During the network training stage, we employ a multi-gather simultaneous inversion method, combining one-dimensional and two-dimensional inversions. The proposed method is named as IGIT (I for Improved ASPP, G for GRU, I for Initial model and T for Triplet attention). To verify the feasibility of this network model, we evaluate it using the Marmousi2 model, SEAM model, and field data, comparing the results with other deep learning methods. Experimental results demonstrate that the IGIT not only improves lateral continuity but also delivers accurate and clear inversion details. Notably, the inversion effect for density parameters shows significant enhancement.

Deep Learning Transient Electromagnetic Inversion for Seawater Intrusion

Abstract To enhance the capability of transient electromagnetic method(TEM) in detecting seawater intrusion and delineating the boundaries in coastal areas, we developed a deep learning inversion method for TEM data based on the Swin Transformer model in this study. First many standardized resistivity models were designed and generated to describe t the subsurface resistivity structures associated with seawater intrusion in coastal areas .Then, TEM forward modelling was performed to compute the corresponding TEM responses, thereby constructing a seawater intrusion-oriented training dataset. Then, the robust Swin Transformer model was employed as the backbone network to build a deep learning inversion model, named SITEMNet, to derive a direct nonlinear transformation that maps TEM responses to subsurface resistivity models. The proposed SITEMNet inversion technique was validated using simulated data scenarios and actual field TEM measurements, showing great promise in accurately identifying seawater intrusion interface and geological formations.