Double-scale supervised inversion with a data-driven forward model for low-frequency impedance recovery

Yuan, Sanyi; Wang, Shangxu; Sang, Wenjing; Jiao, Xiaohong; Luo, Yaneng

doi:10.1190/geo2020-0421.1

Cited by 41 publications

(8 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Adding the constraint of low-frequency details can speed up the convergence of inversion and improve the accuracy of inversion (Yuan et al, 2019). If the low-frequency feature information is missing, the impedance inversion results will be biased, which may lead to inversion results not accurately reflecting stratigraphic changes (Yuan et al, 2022). (Biswas et al, 2019) applied the low-frequency model to the inversion to extract low-frequency feature information of impedance.…”

Section: Introductionmentioning

confidence: 99%

Low-frequency constrained seismic impedance inversion combining large kernel attention and long short-term memory

Wei,

Li,

Ning

et al. 2024

Acta Geophys.

View full text Add to dashboard Cite

In the seismic impedance inversion, the low-frequency information reflects the general trend of the impedance curve. Without low-frequency information, inversion results can not accurately reflect stratigraphic changes. Seismic data are also spatially correlated, while the conventional inversion methods do not consider the spatial correlation of geological structures, which may lead to poor lateral continuity of the inversion results. To alleviate these problems, we proposes a low-frequency constrained seismic impedance inversion method combining Large Kernel Attention (LKA) and Long Short-Term Memory (LSTM). Our network structure is divided into an inversion module and a low-frequency feature extraction module. In the inversion module, we integrate LKA and LSTM into the network, which can improve the lateral continuity of the inversion results. The low-frequency feature extraction module constrains the entire network structure and extract more refined low-frequency features. To demonstrate the reliability of the proposed method, we applied it to the SEAM model. Experiments show that our method has the best lateral continuity and accuracy, with Mean Squared Error (MSE) and Coefficient of Determination (R 2 ) of 0.0485 and 0.9164, respectively, as well as strong noise immunity. This method also achieves favorable inversion results on the Volve field seismic data.

show abstract

Section: Introductionmentioning

confidence: 99%

Low-frequency constrained seismic impedance inversion combining large kernel attention and long short-term memory

Wei,

Li,

Ning

et al. 2024

Acta Geophys.

View full text Add to dashboard Cite

show abstract

“…Each hidden layer of the recurrent neural networks (RNNs) has a feedback to a previous layer, and the subsequent behavior can be shaped by the response of the previous layer. Thus, RNNs are well suited for processing sequential data, and since logging data are connected indepth, RNNs and their variants long short-term memory (LSTM) networks and gated recurrent units (GRU) networks have been introduced into the S-wave velocity prediction (Mehrgini et al, 2017;Zhang et al, 2020) and other rock parameters (Yuan et al, 2022). Moreover, convolutional neural networks (CNNs) have tremendous advantages in feature extraction, thus the CNNs were widely developed and applied in many research fields (Yuan et al, 2018;Hu et al, 2020;Hu et al, 2021), and a combination of RNNs and CNNs for S-wave velocity prediction has been proposed recently (Wang et al, 2022;Zhang et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Shear wave velocity prediction based on deep neural network and theoretical rock physics modeling

Feng

Zeng²,

Xu³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

Shear wave velocity plays an important role in both reservoir prediction and pre-stack inversion. However, the current deep learning-based shear wave velocity prediction methods have certain limitations, including lack of training dataset, poor model generalization, and poor physical interpretability. In this study, the theoretical rock physics models are introduced into the construction of the labeled dataset for deep learning algorithms, and a forward simulation of the theoretical rock physics models is utilized to supplement the dataset that incorporates geological and geophysical knowledge. This markedly increases the physical interpretability of the deep learning algorithm. Theoretical rock physics models for two different types of reservoirs, i.e., conventional sandstone and tight sandstone reservoirs, are first established. Then, a full-sample labeled dataset is constructed using these two types of theoretical rock physics models to traverse the elasticity parameter space of the two types of reservoirs through random variation and combination of parameters in the theoretical models. Finally, based on the constructed full-sample labeled dataset, four parameters (P-wave velocity, clay content, porosity, and density) that are highly correlated with the shear wave velocity are selected and combined with a deep neural network to build a deep shear wave velocity prediction network with good generalization and robustness, which can be directly applied to field data. The errors between the predicted shear wave velocity using the deep neural network and the measured shear wave velocity data in the laboratory and the logging data in three real field work areas are less than 5%, which are much smaller than the errors predicted by both Han’s and Castagna’s empirical formula. Furthermore, the prediction accuracy and generalization performance are better than those of these two common empirical formulas. The forward simulation based on theoretical models supplements the training dataset and provides high-quality labels for machine learning. This can considerably improve the interpretability and generalization of models in real applications of a machine learning algorithm.

show abstract

“…Therefore, many fault detection methods are proposed to enhance those discontinuities using some seismic attributes including the semblance, coherence and curvature (Marfurt et al, 1998;Marfurt et al, 1999;Roberts, 2001). To pursue better performance, more improved approaches are proposed including the ant tracking and attributes fused methods (Pedersen et al, 2002;Di et al, 2019;Yuan et al, 2020;Acuña-Uribe et al, 2021;Yuan et al, 2022), but the results still rely heavily on the experience of interpreters and the quality of the seismic attributes used. Moreover, the presence of noise in seismic images can negatively impact the accuracy of fault detection.…”

Section: Introductionmentioning

confidence: 99%

Transformer assisted dual U-net for seismic fault detection

et al. 2023

View full text Add to dashboard Cite

Automatic seismic fault identification for seismic data is essential for oil and gas resource exploration. The traditional manual method cannot accommodate the needs of processing massive seismic data. With the development of artificial intelligence technology, deep learning techniques based on pattern recognition have become a popular research area for seismic fault identification. Despite the progress made with U-shaped neural networks (Unet), they still fall short in meeting the stringent requirements of fault prediction in complex structures. We propose a novel approach by combining a standard Unet with a transformer Unet to create a parallel dual Unet model, called Dual Unet with Transformer. To improve the accuracy of fault prediction, we compare six loss functions (including Binary Cross Entropy loss, Dice coefficient loss, Tversky loss, Local Tversky loss, Multi-scale Structural Similarity and Intersection over Union loss) using synthetic data, based on three evolution metrics involving Dice coefficient, Sensitivity and Specificity, find that the binary cross entropy loss function is the most robust one. An example comparing the prediction performance of different Unet models on synthetic data demonstrates the superior performance of our Dual Unet model, verifying the practical application value. To further validate the practical feasibility of our proposed method, we use real seismic data with a complex fault system and find that our proposed model is more accurate in predicting the fault system compared to well-developed Unet models such as the classical Unet and classical coherence cube algorithm, without transfer learning. This confirms the potential for wide-scale application of our proposed model.

show abstract

Double-scale supervised inversion with a data-driven forward model for low-frequency impedance recovery

Cited by 41 publications

References 58 publications

Low-frequency constrained seismic impedance inversion combining large kernel attention and long short-term memory

Low-frequency constrained seismic impedance inversion combining large kernel attention and long short-term memory

Shear wave velocity prediction based on deep neural network and theoretical rock physics modeling

Transformer assisted dual U-net for seismic fault detection

Contact Info

Product

Resources

About