Prediction of Shear Wave Velocity Based on a Hybrid Network of Two-Dimensional Convolutional Neural Network and Gated Recurrent Unit

Abstract

Compressional and shear wave velocities (Vp and Vs, respectively) are important elastic parameters to predict reservoir parameters, such as lithology and hydrocarbons. Due to acquisition technologies and economy, the shear wave velocity is generally lacking. Over the last few years, some researchers proposed deep learning algorithms to predict the shear wave velocity using conventional logging data. However, these algorithms focus either on spatial feature extraction for different physical properties of rocks or on sequential feature extraction in the depth direction of rocks. Only focusing on feature extraction in a direction of rocks might lead to a decrease in prediction accuracy. Therefore, we propose a hybrid network of a two-dimensional convolutional neural network and the gated recurrent unit (2DCNN-GRU), which can establish more complex nonlinear relationships between the input and output data based on the spatial features extracted by 2DCNN and the sequential features extracted by GRU. In this study, two cases are used to validate the reliability and prediction accuracy of the proposed network. Comparing the prediction results of 2DCNN, GRU, and the proposed network, the proposed network shows better performance. Meanwhile, for improving the prediction accuracy of the proposed network, the relationship is analyzed between the prediction accuracy of the proposed network and the length of the input sample.

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

“…Shear wave velocity is an important parameter in reservoir prediction tasks and is commonly used for reservoir lithology, reservoir physical properties, and reservoir fluid identification and prediction (Russell et al, 2003;Rezaee et al, 2007;Oloruntobi and Butt, 2020). Accurate logging of S-wave velocities is also helpful to prestack seismic inversion and prestack seismic attribute analysis, however, due to the high cost of S-wave logging, S-wave velocities are often not available in many field areas, especially in old wells (Bagheripour et al, 2015;Chen T. et al, 2022). To address this problem, many scholars have proposed three categories of methods for S-wave velocity prediction, including using an empirical formula, rock physics modeling, and from machine learning predictions.…”

Shear Wave Velocity Prediction Based on the Long Short-Term Memory Network with Attention Mechanism

“…Compared with the ANN, which only considers the petrophysical characteristics at the same depth, LSTM synthesizes the contextual information of logging data to predict shear wave velocity, achieving more accurate prediction results [38,39]. In addition, many scholars have integrated the powerful feature extraction capability of convolutional neural networks (CNNs) into the LSTM network and achieved good results on shear wave prediction [40,41]. Considering the sequence characteristics of log data, the LSTM network can increase the ability of the feedforward neural network to characterize log data under different conditions by introducing recurrent structures.…”