Abstract
The emerging applications of deep learning in solving geophysical problems have attracted increasing attention. In particular, it is of significance to enhance the computational efficiency of the computationally intensive geophysical algorithms. In this paper, we accelerate deabsorption prestack time migration (QPSTM), which can yield higher-resolution seismic imaging by compensating absorption and correcting dispersion through deep learning. This is implemented by training a neural network with pairs of small-sized patches of the stacked migrated results obtained by conventional PSTM and deabsorption QPSTM and then yielding the high-resolution imaging volume by prediction with the migrated results of conventional PSTM. We use an encoder-decoder network to highlight the features related to high-resolution migrated results in a high-order dimension space. The training data set of small-sized patches not only reduces the required high-resolution migrated result (for instance, only several inline is required) but leads to a fast convergence in training. The proposed deep-learning approach accelerates the high-resolution imaging by more than 100 times. Field data is used to demonstrate the effectiveness of the proposed method.
Highlights
Understanding the underground structure is essential to energy exploration, avoiding natural disasters, and studying the evolution of the Earth
The use of the graphics processing units (GPUs) is beneficial, the approach remains less efficient for solving realistic problems; for instance, performing a realistic 3D QPSTM can take several months even when using advanced
The corresponding high-resolution imaging patches have more detailed information. These patches are fed into the network, and the network is trained using the TensorFlow program [39] by employing a GPU
Summary
Understanding the underground structure is essential to energy exploration, avoiding natural disasters, and studying the evolution of the Earth. Pre-stack time migration (PSTM) has been widely used in academia and industry, mainly because of its robustness against velocity errors. The conventional PSTM ignores the viscosity of the medium, which leads to the absence of high-frequency components and results in low-resolution migration images. Zhang [1,2] introduced an effective Q approach (QPSTM). To compensate for the attenuated energies, which are crucial for high-resolution imaging. In contrast to the conventional PSTM, the QPSTM approach involves performing an additional integral over all the frequencies, which is time consuming and prevents the application of this approach to large-scale problems. The use of the GPUs is beneficial, the approach remains less efficient for solving realistic problems; for instance, performing a realistic 3D QPSTM can take several months even when using advanced
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call