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

Abstract

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.

Comparing the performance of machine learning methods in estimating the shear wave transit time in one of the reservoirs in southwest of Iran

“…The DNN model was very accurate in velocity prediction, with errors of less than 5% in both laboratory and field domains. The findings of this study highlight the potential of using DNN algorithms to estimate subsurface features 24 .…”

Research on prediction of slurry migration distance in aggregate stacking based on GA-PSO-BPNN algorithm

“…Slurry injection pressure, water cement ratio, and aggregate stacking porosity are selected from the table as input influencing factors for the neural network, with the lateral migration distance of the slurry as the target output data. The data must be normalized to avoid prediction errors in the neural network due to significant differences in the magnitudes of input and output data (Feng et al, 2022;Shan et al, 2022). The data set is normalized to the range [0, 1] using the mapminmax function provided by Matlab.…”

Shear wave velocity prediction based on 1DCNN-BiLSTM network with attention mechanism

“…Machine learning (ML) algorithms and neural networks have an advantage in extracting relationships between various data (Thanh et al, 2024a;Thanh et al, 2024b;Ewees et al, 2024;Zhang et al, 2024), which can serve in establishing an accurate nonlinear relationship between S-wave velocity and reservoir parameters. Therefore, the prediction of S-wave velocity using logging data and neural networks has been widely employed in field data (Alimoradi et al, 2011;Maleki et al, 2014;Mehrgini et al, 2017;Feng et al, 2023). However, conventional neural networks only establish a point-to-point relationship between logging data and S-wave velocity, without considering the variation pattern of the logging curve at depth, resulting in limited accuracy of S-wave velocity prediction.…”