scholarly journals Seismic inversion by Newtonian machine learning

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. WA185-WA200
Author(s):  
Yuqing Chen ◽  
Gerard T. Schuster
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Wave Equation ◽  
Seismic Data ◽  
Velocity Model ◽  
Seismic Inversion ◽  
Low Rank ◽  
Inversion Method ◽  
Model Parameters ◽  
Latent Space

We present a wave-equation inversion method that inverts skeletonized seismic data for the subsurface velocity model. The skeletonized representation of the seismic traces consists of the low-rank latent-space variables predicted by a well-trained autoencoder neural network. The input to the autoencoder consists of seismic traces, and the implicit function theorem is used to determine the Fréchet derivative, i.e., the perturbation of the skeletonized data with respect to the velocity perturbation. The gradient is computed by migrating the shifted observed traces weighted by the skeletonized data residual, and the final velocity model is the one that best predicts the observed latent-space parameters. We denote this as inversion by Newtonian machine learning (NML) because it inverts for the model parameters by combining the forward and backward modeling of Newtonian wave propagation with the dimensional reduction capability of machine learning. Empirical results suggest that inversion by NML can sometimes mitigate the cycle-skipping problem of conventional full-waveform inversion (FWI). Numerical tests with synthetic and field data demonstrate the success of NML inversion in recovering a low-wavenumber approximation to the subsurface velocity model. The advantage of this method over other skeletonized data methods is that no manual picking of important features is required because the skeletal data are automatically selected by the autoencoder. The disadvantage is that the inverted velocity model has less resolution compared with the FWI result, but it can serve as a good initial model for FWI. Our most significant contribution is that we provide a general framework for using wave-equation inversion to invert skeletal data generated by any type of neural network. In other words, we have combined the deterministic modeling of Newtonian physics and the pattern matching capabilities of machine learning to invert seismic data by NML.

Traveltime information-based wave-equation inversion

Geophysics ◽  
2009 ◽  
Vol 74 (6) ◽  
pp. WCC27-WCC36 ◽  
Author(s):  
Yu Zhang ◽  
Daoliu Wang
Keyword(s):  
Wave Equation ◽  
Conventional Method ◽  
Seismic Data ◽  
Waveform Inversion ◽  
Back Propagation ◽  
Velocity Model ◽  
Data Fitting ◽  
Inversion Method ◽  
New Wave ◽  
Seismic Record

We propose a new wave-equation inversion method that mainly depends on the traveltime information of the recorded seismic data. Unlike the conventional method, we first apply a [Formula: see text] transform to the seismic data to form the delayed-shot seismic record, back propagate the transformed data, and then invert the velocity model by maximizing the wavefield energy around the shooting time at the source locations. Data fitting is not enforced during the inversion, so the optimized velocity model is obtained by best focusing the source energy after a back propagation. Therefore, inversion accuracy depends only on the traveltime information embedded in the seismic data. This method may overcome some practical issues of waveform inversion; in particular, it relaxes the dependency of the seismic data amplitudes and the source wavelet.

THE G-LOG* SEISMIC INVERSION PROCESS

1981 ◽  
Vol 21 (1) ◽  
pp. 155
Author(s):  
D. B. Hays ◽  
J. Wardell
Keyword(s):  
Wave Equation ◽  
Seismic Data ◽  
Acoustic Impedance ◽  
Mean Squared Error ◽  
Seismic Inversion ◽  
Synthetic Seismogram ◽  
Inversion Method ◽  
Multiple Reflections ◽  
Seismic Section ◽  
Impedance Model

The G-LOG process is a method of seismic inversion which provides direct estimates of subsurface acoustic impedance from wavelet process stacked or migrated data. The fundamentals and characteristics of the inversion method will be discussed and examples of its use on Australian seismic data will be presented.G-LOG functions are derived by an iterative subsurface modelling technique based on a rigorous inversion of one- dimensional wave equation. This process finds the acoustic impedance model, or log, whose resulting wave-equation- consistent synthetic seismogram best matches the input seismic data in a least mean squared error sense. Multiple reflections are included in the synthetic seismogram, so that they become useful information in the determination of the log.Interval velocity logs are derived from the acoustic impedance logs. The results can be displayed in various forms, including detailed velocity logs, and colour-coded log 'sections' to match with the seismic section. Several examples of such results are presented.The G-LOG process is a revolutionary technique of subsurface modelling, and the logs it provides are strong indicators of subsurface lithology and will be an important tool in the evaluation and re-evaluation of potential hydrocarbon-bearing prospects.*Trademark of G.S.I.

Thermal–Mechanical Coupling Evaluation of the Panel Performance of a Prefabricated Cabin-Type Substation Based on Machine Learning

Fire ◽  
2021 ◽  
Vol 4 (4) ◽  
pp. 93
Author(s):  
Xiangsheng Lei ◽  
Jinwu Ouyang ◽  
Yanfeng Wang ◽  
Xinghua Wang ◽  
Xiaofeng Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Numerical Simulation ◽  
Fire Resistance ◽  
Bp Neural Network ◽  
Stress Test ◽  
Normal Operation ◽  
Model Parameters ◽  
Mechanical Coupling ◽  
The Impact

The panel performance of a prefabricated cabin-type substation under the impact of fires plays a vital role in the normal operation of the substation. However, current evaluations of the panel performance of substations under fire still focus on fire resistance tests, which seldom consider the relationship between fire behavior and the mechanical load of the panel under the impact of fires. Aiming at the complex and uncertain relationship between the thermal and mechanical performance of the substation panel under impact of fires, this paper proposes a machine learning method based on a BP neural network. First, the fire resistance test and the stress test of the panel is carried out, then a machine learning model is established based on the BP neural network. According to the collected data, the model parameters are obtained through a series of training and verification processes. Meanwhile, the correlation between the panel performance and fire resistance was obtained. Finally, related parameters are input into the thermal–mechanical coupling evaluation model for the substation panel performance to evaluate the fire resistance performance of the substation panel. To verify the correctness of the established model, numerical simulation of the fire test and stress test of the panel is conducted, and numerical simulation samples are predicted by the trained model. The results show that the prediction curve of neural network is closer to the real results compared with the numerical simulation, and the established model can accurately evaluate the thermal–mechanical coupling performance of the substation panel under fire.

Unravel Carbonate Reservoir Heterogeneities Through 3D Seismic Neural Network Application: A Machine Learning Based Approach and Workflow

2021 ◽  
Author(s):  
Dimmas Ramadhan ◽  
Krishna Pratama Laya ◽  
Ricko Rizkiaputra ◽  
Esterlinda Sinlae ◽  
Ari Subekti ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Seismic Data ◽  
Carbonate Reservoir ◽  
3D Seismic ◽  
Carbonate Facies ◽  
Full Field ◽  
Facies Classification ◽  
Unsupervised Neural Network ◽  
Tight Gas Reservoir

Abstract The availability of 3D seismic data undoubtedly plays an important role in reservoir characterization. Currently seismic technology continues to advance at a rapid pace not only in the acquisition but also in processing and interpretation domain. The advance on this is well supported by the digitalization era which urges everything to run reliably fast, effective and efficient. Thanks to continuous development of IT peripherals we now have luxury to process and handle big data through the application of machine learning. Some debates on the effectiveness and threats that this process may automating certain task and later will decrease human workforce are still going on in many forums but still like it or not this machine learning is already embraced in almost every aspect of our life including in oil & gas industry. Carbonate reservoir on the other hand has been long known for its uniqueness compared to siliciclastic reservoir. The term heterogeneous properties are quite common for carbonate due to its complex multi-story depositional and diagenetic facies. In this paper, we bring up our case where we try to unravel carbonate heterogeneity from a massive tight gas reservoir through our machine learning application using the workflow of supervised and unsupervised neural network. In this study, we incorporate 3D PSTM seismic data and its stratigraphic interpretation coupled with the core study result, BHI (borehole image) log interpretation, and our regional understanding of the area to develop a meaningful carbonate facies model through seismic neural network exercises. As the result, we successfully derive geological consistent carbonate facies classification and distribution honoring all the supporting data above though the limitation of well penetration in the area. This result then proved to be beneficial to build integrated 3D geomodel which later can explain the issue on different gas compositions happens in the area. The result on unsupervised neural network also able to serves as a quick look for further sweetspot analysis to support full-field development.

A New Inversion Method Using a Modified Bat Algorithm for Analysis of Seismic Refraction Data in Dam Site Investigation

2019 ◽  
Vol 24 (2) ◽  
pp. 201-214
Author(s):  
Rashed Poormirzaee ◽  
Siamak Sarmady ◽  
Yusuf Sharghi
Keyword(s):  
Seismic Data ◽  
Seismic Refraction ◽  
Bat Algorithm ◽  
Inversion Method ◽  
Model Parameters ◽  
Inversion Algorithm ◽  
Dam Site ◽  
Seismic Refraction Data ◽  
Modified Bat Algorithm ◽  
Refraction Data

Similar to any other geophysical method, seismic refraction method faces non-uniqueness in the estimation of model parameters. Recently, different nonlinear seismic processing techniques have been introduced, particularly for seismic inversion. One of the recently developed metaheuristic algorithms is bat optimization algorithm (BA). Standard BA is usually quick at the exploitation of the solution, while its exploration ability is relatively poor. In order to improve exploration ability of BA, in the current study, a hybrid metaheuristic algorithm by inclusion a mutation operator into BA, so-called mutation based bat algorithm (MBA), is introduced to inversion of seismic refraction data. The efficiency and stability of the proposed inversion algorithm were tested on different synthetic cases. Finally, the MBA inversion algorithm was applied to a real dataset acquired from Leylanchay dam site at East-Azerbaijan province, Iran, to determine alluvium depth. Then, the performance of MBA on both synthetic and real datasets was compared with standard BA. Moreover, the dataset was further processed following a tomographic approach and the results were compared to the results of the proposed MBA inversion method. In general, the MBA inversion results were superior to standard BA inversion and results of MBA were in good agreement with available boreholes data and geological sections at the dam site. The analysis of the seismic data showed that the studied site comprises three distinct layers: a saturated alluvial, an unsaturated alluvial, and a dolomite bedrock. The measured seismic velocity across the dam site has a range of 400 to 3,500 m/s, with alluvium thickness ranging from 5 to 19 m. Findings showed that the proposed metaheuristic inversion framework is a simple, fast, and powerful tool for seismic data processing.

A comparison of deep machine learning and Monte Carlo methods for facies classification from seismic data

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. WA41-WA52 ◽  
Author(s):  
Dario Grana ◽  
Leonardo Azevedo ◽  
Mingliang Liu
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Monte Carlo ◽  
Monte Carlo Methods ◽  
Seismic Data ◽  
Transition Probabilities ◽  
Machine Learning Algorithms ◽  
Data Set ◽  
Facies Classification ◽  
The Neural Network

Among the large variety of mathematical and computational methods for estimating reservoir properties such as facies and petrophysical variables from geophysical data, deep machine-learning algorithms have gained significant popularity for their ability to obtain accurate solutions for geophysical inverse problems in which the physical models are partially unknown. Solutions of classification and inversion problems are generally not unique, and uncertainty quantification studies are required to quantify the uncertainty in the model predictions and determine the precision of the results. Probabilistic methods, such as Monte Carlo approaches, provide a reliable approach for capturing the variability of the set of possible models that match the measured data. Here, we focused on the classification of facies from seismic data and benchmarked the performance of three different algorithms: recurrent neural network, Monte Carlo acceptance/rejection sampling, and Markov chain Monte Carlo. We tested and validated these approaches at the well locations by comparing classification predictions to the reference facies profile. The accuracy of the classification results is defined as the mismatch between the predictions and the log facies profile. Our study found that when the training data set of the neural network is large enough and the prior information about the transition probabilities of the facies in the Monte Carlo approach is not informative, machine-learning methods lead to more accurate solutions; however, the uncertainty of the solution might be underestimated. When some prior knowledge of the facies model is available, for example, from nearby wells, Monte Carlo methods provide solutions with similar accuracy to the neural network and allow a more robust quantification of the uncertainty, of the solution.

Target-oriented wave-equation inversion

Geophysics ◽  
2006 ◽  
Vol 71 (4) ◽  
pp. A35-A38 ◽  
Author(s):  
Alejandro A. Valenciano ◽  
Biondo Biondi ◽  
Antoine Guitton
Keyword(s):  
Wave Equation ◽  
Least Squares ◽  
Seismic Data ◽  
Hessian Matrix ◽  
Velocity Model ◽  
Complex Model ◽  
Inverse Filtering ◽  
Band Limited ◽  
Filtering Problem ◽  
Number Of Iterations

A target-oriented strategy can be applied to estimate a wave-equation least-squares inverse (LSI) image. By explicitly computing the wave-equation Hessian, the LSI image is obtained as the solution of a nonstationary least-squares inverse filtering problem. The rows of the Hessian are the nonstationary filters containing information about the acquisition geometry, the velocity model, and the band-limited characteristics of the seismic data. By exploiting the sparsity and the structure of the Hessian matrix, a large number of iterations, necessary to achieve convergence, can be computed cheaply. The results on a structurally complex model show the improvements of the LSI image versus the migrated image.

Accuracy of wavelets, seismic inversion, and thin-bed resolution

2017 ◽  
Vol 5 (4) ◽  
pp. T523-T530
Author(s):  
Ehsan Zabihi Naeini ◽  
Mark Sams
Keyword(s):  
Seismic Data ◽  
Reservoir Characterization ◽  
Low Frequency ◽  
Seismic Inversion ◽  
Inversion Method ◽  
Frequency Content ◽  
Frequency Model ◽  
North West ◽  
The North ◽  
Well Data

Broadband reprocessed seismic data from the North West Shelf of Australia were inverted using wavelets estimated with a conventional approach. The inversion method applied was a facies-based inversion, in which the low-frequency model is a product of the inversion process itself, constrained by facies-dependent input trends, the resultant facies distribution, and the match to the seismic. The results identified the presence of a gas reservoir that had recently been confirmed through drilling. The reservoir is thin, with up to 15 ms of maximum thickness. The bandwidth of the seismic data is approximately 5–70 Hz, and the well data used to extract the wavelet used in the inversion are only 400 ms long. As such, there was little control on the lowest frequencies of the wavelet. Different wavelets were subsequently estimated using a variety of new techniques that attempt to address the limitations of short well-log segments and low-frequency seismic. The revised inversion showed greater gas-sand continuity and an extension of the reservoir at one flank. Noise-free synthetic examples indicate that thin-bed delineation can depend on the accuracy of the low-frequency content of the wavelets used for inversion. Underestimation of the low-frequency contents can result in missing thin beds, whereas underestimation of high frequencies can introduce false thin beds. Therefore, it is very important to correctly capture the full frequency content of the seismic data in terms of the amplitude and phase spectra of the estimated wavelets, which subsequently leads to a more accurate thin-bed reservoir characterization through inversion.

Probabilistic petrophysical-properties estimation integrating statistical rock physics with seismic inversion

Geophysics ◽  
2010 ◽  
Vol 75 (3) ◽  
pp. O21-O37 ◽  
Author(s):  
Dario Grana ◽  
Ernesto Della Rossa
Keyword(s):  
Seismic Data ◽  
Measurement Errors ◽  
Rock Physics ◽  
Seismic Inversion ◽  
Joint Estimation ◽  
Physical Models ◽  
Inversion Method ◽  
Petrophysical Properties ◽  
Reservoir Properties ◽  
Statistical Rock Physics

A joint estimation of petrophysical properties is proposed that combines statistical rock physics and Bayesian seismic inversion. Because elastic attributes are correlated with petrophysical variables (effective porosity, clay content, and water saturation) and this physical link is associated with uncertainties, the petrophysical-properties estimation from seismic data can be seen as a Bayesian inversion problem. The purpose of this work was to develop a strategy for estimating the probability distributions of petrophysical parameters and litho-fluid classes from seismics. Estimation of reservoir properties and the associated uncertainty was performed in three steps: linearized seismic inversion to estimate the probabilities of elastic parameters, probabilistic upscaling to include the scale-changes effect, and petrophysical inversion to estimate the probabilities of petrophysical variables andlitho-fluid classes. Rock-physics equations provide the linkbetween reservoir properties and velocities, and linearized seismic modeling connects velocities and density to seismic amplitude. A full Bayesian approach was adopted to propagate uncertainty from seismics to petrophysics in an integrated framework that takes into account different sources of uncertainty: heterogeneity of the real data, approximation of physical models, measurement errors, and scale changes. The method has been tested, as a feasibility step, on real well data and synthetic seismic data to show reliable propagation of the uncertainty through the three different steps and to compare two statistical approaches: parametric and nonparametric. Application to a real reservoir study (including data from two wells and partially stacked seismic volumes) has provided as a main result the probability densities of petrophysical properties and litho-fluid classes. It demonstrated the applicability of the proposed inversion method.

Case Study: Improving the quality of the seismic reflection image for a Fujian mineral exploration data set with offset-domain common-image gathers

Geophysics ◽  
2021 ◽  
pp. 1-45
Author(s):  
Guofeng Liu ◽  
Xiaohong Meng ◽  
Johanes Gedo Sea
Keyword(s):  
Wave Equation ◽  
Image Quality ◽  
Seismic Data ◽  
Seismic Reflection ◽  
Mineral Exploration ◽  
Velocity Model ◽  
Image Point ◽  
Single Shot ◽  
Data Set ◽  
Depth Migration

Seismic reflection is a proven and effective method commonly used during the exploration of deep mineral deposits in Fujian, China. In seismic data processing, rugged depth migration based on wave-equation migration can play a key role in handling surface fluctuations and complex underground structures. Because wave-equation migration in the shot domain cannot output offset-domain common-image gathers in a straightforward way, the use of traditional tools for updating the velocity model and improving image quality can be quite challenging. To overcome this problem, we employed the attribute migration method. This worked by sorting the migrated stack results for every single-shot gather into the offset gathers. The value of the offset that corresponded to each image point was obtained from the ratio of the original migration results to the offset-modulated shot-data migration results. A Gaussian function was proposed to map every image point to a certain range of offsets. This helped improve the signal-to-noise ratio, which was especially important in handing low quality seismic data obtained during mineral exploration. Residual velocity analysis was applied to these gathers to update the velocity model and improve image quality. The offset-domain common-image gathers were also used directly for real mineral exploration seismic data with rugged depth migration. After several iterations of migration and updating the velocity, the proposed procedure achieved an image quality better than the one obtained with the initial velocity model. The results can help with the interpretation of thrust faults and deep deposit exploration.

Sign in / Sign up

Export Citation Format

Share Document