Conventional Inversion Research Articles

Trans-dimensional gravity and magnetic joint inversion for 3-D earth models

SUMMARY Studying 3-D subsurface structure based on spatial data is an important application for geophysical inversions. However, major limitations exist for conventional regularized inversion when applied to potential-field data. For example, global regularization parameters can mask model features that may be important for interpretation. In addition, 3-D inversions are typically based on data acquired in 2-D at the Earth’s surface. Such data may contain significant spatial error correlations in 2-D due to the choice of spatial sampling, acquisition geometry, ambient noise and model assumptions. These correlations can cause trade-offs with spatial resolution and should be accounted for. However, correlations are often ignored, particularly 2-D correlations in spatial data, such as potential field data recorded on the Earth’s surface. Non-linear Bayesian methods can address these shortcomings and we present a new hierarchical model for 2-D correlated errors. Nonetheless, limitations also exist. For example, non-linear Bayesian estimation requires numerical integration with a considerable computational burden to collect a posterior ensemble of models. For 3-D applications, this cost can be prohibitive. This paper presents a non-linear Bayesian inversion with trans-dimensional (trans-D) partitioning of space by a hierarchy of Voronoi nodes and planes (VP), and trans-D estimation of the data noise covariance matrix. The addition of planes permits the introduction of prior information which reduces non-uniqueness. The covariance matrix estimation uses a trans-D autoregressive (AR) noise model to quantify correlated noise on 2-D potential-field data. We address computational cost by wavelet compression in the forward problem and by basing susceptibility on an empirical relationship with density contrast. The method is applied to simulated data and field data from off-shore Abu Dhabi. With simulated data, we demonstrate that subsurface structures are well-resolved with the trans-D model that applies hierarchical VP partitioning. In addition, the model locally adapts based on data information without requiring regularization. The method is also successful in reducing 2-D error correlation via trans-D AR models in 2-D. From field data, the inversion efficiently resolves basement topography and two distinct salt diapirs with a parsimonious and data-driven parametrization. Results show a considerable reduction in 2-D spatial correlations of field data using the proposed trans-D AR model.

A novel inversion method of 2D TD-NMR signals based on realizing unconstrained maximization of objective function

The inversion of time-domain nuclear magnetic resonance (TD-NMR) signals is an ill-posed problem, which presents enormous challenges for the inversion algorithm. We propose a novel inversion method that converts conventional minimum objective function with non-negative constraints into an unconstrained maximization problem in the inversion of TD-NMR signals. Hence, the objective function becomes a differentiable concave function that can be solved more easily. The validity of the proposed method was verified by the uncertainty estimation of NMR inversion spectra with different signal-to-noise ratios (SNR). Through the inversion of simulated 2D D-T2 and T1-T2 signals under different SNR, the proposed method was proved to be less sensitive to noise than the conventional inversion method. We use the proposed method to study the migrations of oil and water in shales, the components change in shale could be identified and quantified according to the 2D T1-T2 inversion spectra. The proposed method was also used to analyze the hydration process of cement. The 2D T1-T2 inversion spectra could distinctly present the component of tiny volume with short relaxation time, and the migration regularity of capillary water, gel water, and bound water could also be found. In conclusion, the proposed method could be a reliable method to invert TD-NMR signals, especially the identification of the 2D NMR signals with a short relaxation time in low SNR.

Model Regeneration Scheme Using a Deep Learning Algorithm for Reliable Uncertainty Quantification of Channel Reservoirs

Abstract Reservoir characterization is one of the essential procedures for decision makings. However, conventional inversion methods of history matching have several inevitable issues of losing geological information and poor performances, when it is applied to channel reservoirs. Therefore, we propose a model regeneration scheme for reliable uncertainty quantification of channel reservoirs without conventional model inversion methods. The proposed method consists of three parts: feature extraction, model selection, and model generation. In the feature extraction part, drainage area localization and discrete cosine transform are adopted for channel feature extraction in near-wellbore area. In the model selection part, K-means clustering and an ensemble ranking method are utilized to select models that have similar characteristics to a true reservoir. In the last part, deep convolutional generative adversarial networks (DCGAN) and transfer learning are applied to generate new models similar to the selected models. After the generation, we repeat the model selection process to select final models from the selected and the generated models. We utilize these final models to quantify uncertainty of a channel reservoir by predicting their future productions. After applying the proposed scheme to three different channel fields, it provides reliable models for production forecasts with reduced uncertainty. The analyses show that the scheme can effectively characterize channel features and increase a probability of existence of models similar to a true model.

A Metal-organic composite ultrafiltration membrane synthesized via Quadratic phase inversion

• Stainless steel wire mesh as substrate was used to fabricate composite UF membrane. • A novel strategy denoted as“Quadratic Phase Inversion (QPI)” was proposed. • Composite UF membrane showed high pure water flux and desirable BSA rejection. • Composite UF membrane prepared with “QPI” showed superior running stability. According to the Pore-flow model describing the mass transfer process of Ultrafiltration (UF) membranes, the thickness of the separation layer directly determines the separation performance of an UF membrane: pure water flux (PWF) and foulant rejection. However, the flexibility and limited supporting capacity of the conventional supporting layer (non-woven fabric) limits the further reduction of the separation layer thickness. In this study, stainless steel wire mesh (SSWM) was employed as the supporting layer to fabricate a metal/polymer composite UF membrane via a novel strategy denoted as “Quadratic Phase Inversion (QPI)”, in which the SSWM was firstly modified with a casting solution with low PVDF concentration, and then another PVDF selective layer was constructed on the modified substrate with the conventional phase inversion method. Orthogonal experiment was conducted to optimize the fabrication parameters. A series of characterization was adopted to analyze the performance and structural evolution. The results demonstrate that a composite UF membrane with ultrathin PVDF selective layer could be fabricated on the PVDF modified SSWM. The optimized sample was obtained with the SSWM modification reagent of 3 wt% PVDF, the casting solution of 20 wt% PVDF, the casting thickness of 75 μm, and NMP as solvent, which had a MWCO of 149 kDa and mean pore diameter of 7.2 nm. Meanwhile, it presented a high PWF around 448.8 L m −2 h −1 bar −1 , desirable BSA rejection about 96.3% and good operation stability. The outcome of this research is expected to provide a valuable route for promoting the separation performance of UF membranes.

Simultaneous inversion of spectral IP data with frequency constraints

Spectral-induced polarization (SIP) measures the apparent complex resistivities (CRs) of electrical current sources at frequencies of less than 10 kHz. The measured apparent CRs are usually inverted to recover the intrinsic subsurface CRs, which depend on the source frequencies. The Cole-Cole (CC) model is commonly used to describe the dispersion characteristics of the intrinsic CRs; electrical resistivity, chargeability, and the time and relaxation constants can be derived using this model. For quantitative estimation of these parameters, a reliable inversion method for multifrequency SIP datasets is required. Conventionally, SIP data at each frequency are independently and sequentially inverted. This does not consider the relationships between SIP datasets of different frequencies. To consider such relationships, here, we simultaneously invert multi-frequency SIP datasets by introducing frequency-smoothness constraints between adjacent frequencies. We numerically verified the utility of the approach in terms of both amplitude and phase recovery; the method yielded more reliable relaxation models (e.g., CC models) than conventional inversion. For field verification, the method was applied to SIP data of several frequencies obtained in South Korea.

Multichannel impedance inversion in the frequency domain via anisotropic total variation with overlapping group sparsity regularization

Abstract Total variation regularization is widely used in geophysical inverse problems for preserving the edges of geological structures. However, it may lead to staircase artifacts and other inversion errors. To overcome this drawback, we present an impedance inversion method by using anisotropic total variation with overlapping group sparsity regularization. This method can not only preserve edges very well but also makes use of the group sparsity feature of acoustic impedance to reduce inversion error. Moreover, we make full use of the properties of the Toeplitz matrix and speed up the inversion process in the frequency domain. The synthetic data and field data inversion results show that the proposed method has the advantages of stronger anti-noise, faster inversion speed, and less error comparing with the conventional total variation regularization inversion method. In addition, our method can effectively alleviate staircase artifacts and therefore improves resolution.

Quality evaluation of glutenite reservoirs of intergranular - secondary dissolution pore type using a quantitative method

Glutenite (coarse-grained clastic) reservoirs of intergranular–secondary dissolution pore type are dominated by residual intergranular pores and secondary dissolution pores, and characterized by low porosity, low permeability, strong heterogeneity, and highly variable physical properties. It is difficult to conduct a quantitative quality assessment of these reservoirs while their primary control factors remain unclear. In this paper, experimental core data and drilling, logging and seismic data are used to assess the effect of sedimentary facies on reservoir quality. Favorable sedimentary facies zones are identified by analyzing the characteristics of glutenite reservoirs, which includes investigating rock components and their effects on reservoir quality. Argillaceous matrix content and rigid particle content are identified as the primary control factors for these reservoirs. Logging curves sensitive to reservoir quality are selected and examined to continuously characterize the physical parameters of the reservoirs. It establishes a calculation model of reservoir assessment parameters through multivariate regression and determines the quantitative assessment parameter Fr. The quality of the glutenite reservoirs is defined using conventional logging curves. This study also predicts the plane distribution of high-quality reservoirs through geostatistical inversion of the reservoir assessment parameters based on conventional wave impedance inversion, thus providing insight and guidance for quantitative assessment and quality prediction of glutenite reservoirs of the intergranular-secondary dissolution pore type. The application of this method to well deployment based on qualitative evaluation of the glutenite reservoirs in oilfields yielded favorable results.

Time-lapse seismic data inversion for estimating reservoir parameters using deep learning

Geologic carbon sequestration involves the injection of captured carbon dioxide ([Formula: see text]) into subsurface formations for long-term storage. The movement and fate of the injected [Formula: see text] plume is of great concern to regulators because monitoring helps to identify potential leakage zones and determines the possibility of safe long-term storage. To address this concern, we design a deep-learning framework for [Formula: see text] saturation monitoring to determine the geologic controls on the storage of the injected [Formula: see text]. We use different combinations of porosities and permeabilities for a given reservoir to generate saturation and velocity models. We train the deep-learning model with a few time-lapse seismic images and their corresponding changes in saturation values for a particular [Formula: see text] injection site. The deep-learning model learns the mapping from the change in the time-lapse seismic response to the change in [Formula: see text] saturation during the training phase. We then apply the trained model to data sets comprising different time-lapse seismic image slices (corresponding to different time instances) generated using different porosity and permeability distributions that are not part of the training to estimate the [Formula: see text] saturation values along with the plume extent. Our algorithm provides a deep-learning assisted framework for the direct estimation of [Formula: see text] saturation values and plume migration in heterogeneous formations using the time-lapse seismic data. Our method improves the efficiency of time-lapse inversion by streamlining the large number of intermediate steps in the conventional time-lapse inversion workflow. This method also helps to incorporate the geologic uncertainty for a given reservoir by accounting for the statistical distribution of porosity and permeability during the training phase. Tests on different examples verify the effectiveness of our approach.

Mapping of hydraulic transmissivity field from inversion of tracer test data using convolutional neural networks. CNN-2T

This paper introduces a new concept for mapping hydraulic transmissivity from temporal concentration data collected in multiple tracer tests. Based on convolutional neural network, the principle uses an encoder-decoder architecture with multiple neural layers to establish a relationship between the concentration data and the transmissivity field. This relationship is established in two phases with two networks. The first network is designated and trained to reconstruct a transmissivity field using data from a single tracer test. To improve the reconstruction quality, the second network then performs a joint interpretation for multiple tracer tests, which reprocesses all the transmissivity resulted from the first network for each individual tracer test. Both networks are trained by synthetic data, where the transmissivity models are generated with a Gaussian variogram and its properties are considered as prior information on the aquifer heterogeneity. Tracer tests are derived numerically by solving the forward problem to obtain the corresponding concentration data that feed the training. The trained networks accurately map the transmissivity fields, of which the accuracy relies on the volume and nature of the heterogeneities of training models, as well as the number of piezometers used to monitor the concentration changes. Reconstruction quality, on the other hand, is less influenced by data noise. Effective training requires a large dataset, but the time required for dataset generation is only on the order of the Gauss-Newton algorithm in a conventional inversion, while the trained network performs inference instantly.

A Multi-Point Geostatistical Seismic Inversion Method Based on Local Probability Updating of Lithofacies

In order to solve the problem that elastic parameter constraints are not taken into account in local lithofacies updating in multi-point geostatistical inversion, a new multi-point geostatistical inversion method with local facies updating under seismic elastic constraints is proposed. The main improvement of the method is that the probability of multi-point facies modeling is combined with the facies probability reflected by the optimal elastic parameters retained from the previous inversion to predict and update the current lithofacies model. Constrained by the current lithofacies model, the elastic parameters were obtained via direct sampling based on the statistical relationship between the lithofacies and the elastic parameters. Forward simulation records were generated via convolution and were compared with the actual seismic records to obtain the optimal lithofacies and elastic parameters. The inversion method adopts the internal and external double cycle iteration mechanism, and the internal cycle updates and inverts the local lithofacies. The outer cycle determines whether the correlation between the entire seismic record and the actual seismic record meets the given conditions, and the cycle iterates until the given conditions are met in order to achieve seismic inversion prediction. The theoretical model of the Stanford Center for Reservoir Forecasting and the practical model of the Xinchang gas field in western China were used to test the new method. The results show that the correlation between the synthetic seismic records and the actual seismic records is the best, and the lithofacies matching degree of the inversion is the highest. The results of the conventional multi-point geostatistical inversion are the next best, and the results of the two-point geostatistical inversion are the worst. The results show that the reservoir parameters obtained using the local probability updating of lithofacies method are closer to the actual reservoir parameters. This method is worth popularizing in practical exploration and development.

3D Bipolar Spectral Inversion Based on Structural Geosteering Shear Backfill Mapping

Conventional multichannel spectral inversion (SI) methods are usually implemented in a 2-D model, which only considers the lateral continuity inside a section and ignores the continuous features between sections in 3-D space. On the other hand, the principle of odd–even decomposition, which can significantly improve the thin-layer recognition ability, is poorly adapted in multichannel with complex structures. We propose a structural geosteering shear backfill mapping (SBM) method to alleviate these issues. After that, we establish a 3-D bipolar SI objective function based on structural geosteering SBM and solve it with 3-D norm regularization. Examples using synthetic and 3-D field data show that the 3-D bipolar SI based on structural geosteering SBM is highly adaptable to complex structures and can provide a better inversion result than the conventional multichannel sparse spike inversion in terms of guaranteeing strata continuities and retrieving weak reflectivity.

Density stability estimation method from pre-stack seismic data

As one of the most important elastic parameters in seismic exploration, density plays an important role in lithology identification and hydrocarbon indicator. Based on the approximate Zoeppritz equation, the density term is usually estimated using AVO/AVA (Amplitude variation with offset/incident angle) inversion. However, the small contribution of density to the reflection coefficient or the strong correlation with other parameters in the approximate equation renders the density estimation tough. By analyzing the contribution of density to reflection coefficient in different approximate equations, Aki-Richards is finally selected as the inversion equation, in which the density term has a higher contribution to the reflectivity. The conventional inversion approach is to invert several parameters simultaneously on the basis of approximate equation. The same convergence criterion is used for the estimation of the three parameters, but the density estimation error fails to converge to the minimum due to its small sensitivity, which makes density inversion unstable. Furthermore, the conventional inversion methods involve the inverse of large matrix, which also reduces the stability of density estimation. Therefore, a new form of Aki-Richards approximate reflection equation is derived on the assumption that the inverse of coefficient matrix exists. The reflectivity of elastic parameter is independently expressed as the weighted superposition of seismic reflection coefficients at different incident angles. Then utilizing well logging data and seismic reflection coefficient estimated by elastic impedance inversion as constraints, the inverse of coefficient matrix in inversion equation is estimated directly, which avoids inverse of large matrix and improves the precision of density estimation. In physical, the weighted coefficients directly reflect the contribution of elastic parameters reflectivity to seismic data. Finally, different experimental examples show that proposed method can stably estimate density term.

Agglomerative clustering to improve the resolution of pseudo well stochastic seismic inversion: A case study

The Monte Carlo rejection and Monte Carlo Markov Chain methods are two common sampling techniques in Bayesian seismic inversion. Conventional methods of pseudo well stochastic seismic inversion for reservoir properties uses Monte Carlo rejection sampling of pseudo-wells to select those that best match seismic traces. The selected pseudo-wells do not influence the selection/rejection of subsequent pseudo-wells. The process requires a very large number of prior pseudo-wells to draw a reasonable outcome, which consumes a major part of the computation time. In conventional methods, the best-matched pseudo-wells often have dissimilarly shaped volume of shale logs, which can degrade the resolution of the outcome. Though Agglomerative clustering is well-known but the idea of grouping pseudo-wells based on similarly shaped volume of shale logs in pseudo well stochastic seismic inversion is unique. We propose an agglomerative clustering technique to group the prior pseudo-wells according to the shape of their volume of shale logs, before matching to seismic. The centroids of the pseudo-well clusters summarize the large number of prior pseudo-wells. The best-matched pseudo-well cluster centroid with seismic gives the most likely outcome while the two next best-matched cluster centroids give alternative results. The alternative results are useful to understand the uncertainty in the outcome. It has been observed that, the proposed method has reduced computation time (50–55% less than conventional pseudo well stochastic inversion). Meanwhile, the quantification of sand volume from the proposed method shows improvement over the conventional method. A comparison between the proposed and the conventional method has been shown on a synthetic wedge model and real seismic. In the wedge model (volume of sand is 0.65), the resultant volume of sand predicted in the modified method (0.7) is better than that predicted using the conventional method (0.8). The modified pseudo-well stochastic seismic inversion method has been applied on seismic data from deepwater Krishna-Godavari basin. The result shows an improvement in the characterization of thin shale layers within the reservoir.

An Efficient Cross-Gradient Joint Inversion Algorithm for Gravity and Magnetic Data Using a Sequential Strategy

The joint inversion methods of gravity and magnetic data have been presented in numerous literatures and implemented in many real applications. However, almost all of these existing methods still face the problem of low computational efficiency, although some targeted improvements have been put forwarded. To further address this issue, this study proposes an efficient algorithm for the joint inversion of gravity and magnetic data. The new algorithm is developed from the idea of structural coupling by using the cross-gradient function and employs a sequential strategy. A new formula for efficient minimization of the structural similarity term is derived innovatively, in which the minimization of the structural coupling term is obtained by a technical combination of an alternating strategy and the randomized singular value decomposition (RSVD) algorithm. Using the proposed new formula, the dimension of the original inverse problem can be reduced greatly. Furthermore, the RSVD algorithm is utilized to perform the independent inversion and determine an optimal regularization parameter. Several comparative tests on a synthetic example show that the proposed efficient method poses great advantage in efficiency compared with the conventional sequential cross-gradient joint inversion (CSCGJI) method. The proposed algorithm is also successfully applied to the real data from a metallic deposit area in Xinjiang province, China.

A new approach for inversion of receiver function for crustal structure in the depth domain

A method for reconstructing crustal velocity structure using the optimization of stacking receiver function amplitude in the depth domain, named common conversion amplitude (CCA) inversion, is presented. The conversion amplitude in the depth domain, which represents the impedance change in the medium, is obtained by assigning the receiver function amplitude to the corresponding conversion position where the P-to-S conversion occurred. Utilizing the conversion amplitude variation with depth as an optimization objective, imposing reliable prior constraints on the structural model frame and velocity range, and adopting a stepwise search inversion technique, this method efficiently weakens the tendency of easily falling into the local extremum in conventional receiver function inversion. Synthetic tests show that the CCA inversion can reconstruct complex crustal velocity structures well and is especially suitable for revealing crustal evolution by estimating diverse velocity distributions. Its performance in reconstructing crustal structure is superior to that of the conventional receiver function imaging method.

Multiscale Data-Driven Seismic Full-Waveform Inversion With Field Data Study

Seismic full-waveform inversion (FWI), which uses iterative methods to estimate high-resolution subsurface models from seismograms, is a powerful imaging technique in exploration geophysics. In recent years, the computational cost of FWI has grown exponentially due to the increasing size and resolution of seismic data. Moreover, it is a non-convex problem and can encounter local minima due to the limited accuracy of the initial velocity models or the absence of low frequencies in the measurements. To overcome these computational issues, we develop a multiscale data-driven FWI method based on fully convolutional networks (FCN). In preparing the training data, we first develop a real-time style transform method to create a large set of synthetic subsurface velocity models from natural images. We then develop two convolutional neural networks with encoder-decoder structure to reconstruct the low- and high-frequency components of the subsurface velocity models, separately. To validate the performance of our data-driven inversion method and the effectiveness of the synthesized training set, we compare it with conventional physics-based waveform inversion approaches using both synthetic and field data. These numerical results demonstrate that, once our model is fully trained, it can significantly reduce the computation time, and yield more accurate subsurface velocity models in comparison with conventional FWI.

Intelligent Dynamic Inversion Controller Design for Ball and Beam System

The Ball and beam system (BBS) is an attractive laboratory experimental tool because of its inherent nonlinear and open-loop unstable properties. Designing an effective ball and beam system controller is a real challenge for researchers and engineers. In this paper, the control design technique is investigated by using Intelligent Dynamic Inversion (IDI) method for this nonlinear and unstable system. The proposed control law is an enhanced version of conventional Dynamic Inversion control incorporating an intelligent control element in it. The Moore-Penrose Generalized Inverse (MPGI) is used to invert the prescribed constraint dynamics to realize the baseline control law. A sliding mode-based intelligent control element is further augmented with the baseline control to enhance the robustness against uncertainties, nonlinearities, and external disturbances. The semi-global asymptotic stability of IDI control is guaranteed in the sense of Lyapunov. Numerical simulations and laboratory experiments are carried out on this ball and beam physical system to analyze the effectiveness of the controller. In addition to that, comparative analysis of RGDI control with classical Linear Quadratic Regulator and Fractional Order Controller are also presented on the experimental test bench.

3-D Inversion of CSEM Data With Hexahedral Mesh in the Multinary Model Space

The controlled-source electromagnetic (CSEM) method is a crucial tool for near-surface investigations and hydrocarbon exploration because of its economic benefits. Limited resolution is one of the inherent defects of the CSEM method, and to obtain high contrast results a multinary transform function was introduced to CSEM 3-D inversion. The multinary transform function was constituted by superposing several error functions, and the transform function transformed the model parameters from continuously distributed space into a semistep distributed multinary space. To deal with complex geometries, the edge-based finite element (FE) method with an irregular hexahedral grid was applied to the modeling and inverse problem. We used the Gauss-Newton optimization method to minimize the Tikhonov parametric function. The complex chessboard model and the marine CSEM model with complex geometry were used to validate the ability of the new method in improving the resolution of the CSEM method. By comparing the results of conventional, focusing, and multinary inversion methods, the effectiveness of the multinary inversion method in depicting sharp boundaries of different physical properties was proved. Additionally, the comparisons proved that multinary inversion method can, to some extent, overcome the insensitivity to low conductors of the CSEM method.

CNN-Based Network Application for Petrophysical Parameter Inversion: Sensitivity Analysis of Input–Output Parameters and Network Architecture

Accurate estimation of petrophysical properties (e.g., porosity, clay volume) of subsurface rock from seismic data/elastic properties is significant to reservoir characterization. Conventional model-driven inversion strategies for estimating petrophysical parameters confront with the deficiency of prior knowledge. In contrast, machine learning-based approaches are adapted to account for reservoir parameter estimation through developing nonlinear mapping and quantifying uncertainty. However, most of the current researches mainly concentrates on the single parameter prediction with different neural network architectures, which, in turn, conflicts with the truth of coupling multiple reservoir properties. To quantify the sensitivity of input-output parameters and the effects of network architecture on the accuracy of petrophysical parameter inversion, we propose a CNN-based network strategy to estimate multiple reservoir parameters simultaneously. The results from both synthetic labeled data and field data and uncertainty analysis strongly demonstrate that SopenCNN, abbreviated from multiple input and single output openCNN, exhibits the highest prediction accuracy, while the cycleCNN with multiple input and multiple output, referred to as McycleCNN, is superior to the MopenCNN, which means that an openCNN contains multiple input and multiple output. It means that, for similar network architecture, the number of input and output parameters makes a significant impact on prediction accuracy. Moreover, for similar multiple inputs and outputs, the fine-tuning McycleCNN, parallelly updated in each intermediate closed-loop step, behaves much better accordingly. The application of the three workflows with varying architecture on field tight sandstone reservoirs demonstrates that network based inversion strategy could establish a mapping function to characterize spatially varying reservoir parameters.

Simultaneous Inversion for Reflectivity and Q Using Nonstationary Seismic Data With Deep-Learning-Based Decoupling

Building reflectivity and quality factor (Q) using nonstationary post-stack seismic data is important for vertical resolution enhancement of seismic data and reservoir identification. However, it is well-known that both reflectivity and Q affect the waveform of seismic data, leading to the fact that simultaneously estimating them is a strong ill-posed multi-parameter inverse problem which faces the crosstalk problem. In this paper, we propose a new method for simultaneous inversion of reflectivity and Q. A deep-learning-based data decoupling operator is proposed to decouple the effects of the two parameters on nonstationary seismic data. Based on the decoupled data, we transform the original multi-parameter inverse problem into two independent singe-parameter inverse problems that are immune to crosstalk and can build reasonable initial models for reflectivity and Q. Then alternative iteration is conducted to update the two built initial models to obtain the final models. A few well-logs are used to train the deep learning architecture and specific regularization terms are constructed for the inverse problem to ensure physically reasonable results. Synthetic and field data examples verify the effectiveness of the proposed method and its advantages over a conventional model-driven joint inversion method.