Seismic Modelling: 4D Capabilities for CO2 Injection

Lubrano-Lavadera, Paul; Drottning, Åsmund; Lecomte, Isabelle; Dando, Ben; Kühn, Daniela; Oye, Volker

doi:10.1016/j.egypro.2017.03.1474

Cited by 4 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods can capture the full range of relevant physics, including the effects of fluid-solid interfaces, intrinsic attenuation, and anisotropy. Both methods estimate the fundamental wave equation to solve for the full seismic wave field propagation [3].…”

Section: Figure 1 Types Of Seismic Wavesmentioning

confidence: 99%

Seismic Waves Near Oil Reservoir Prediction Using Deep Learning

2023

HNSJ

View full text Add to dashboard Cite

Seismic waves, natural events with potential for devastating consequences, necessitate accurate prediction for effective risk assessment and mitigation. Recent advancements have leveraged machine learning to forecast earthquake waves using features like frequency and amplitude. This study introduces a novel feature extraction approach through regression to capture seismic wave characteristics. Long Short-Term Memory (LSTM) and K-Nearest Neighbors (KNR) regression algorithms are then employed to predict earthquake wave properties, like capacity and frequency. The LSTM model shows strong predictive capabilities across most parameters, yielding low RMSE (Root Mean Square Error) and MAE (Mean Absolute Error) values. The KNR model performs well for certain parameters but less consistently across others. Notably, the KNR algorithm's RMSE and MAE values suggest accurate predictions. This method is evaluated using a dataset of seismic recordings from global earthquakes Results underscore the effectiveness of the LSTM algorithm in predicting earthquake wave features. Additionally, this approach outperforms existing methods. Seismic waves, also pertinent in oil fields and mining, have the potential for significant impact. Traditional approaches fall short in modeling spatial relationships, emphasizing the need for modern techniques. Machine learning and deep learning, including LSTM and regression-based feature extraction, offer promising solutions for more accurate and rapid predictions, enhancing safety measures. This research contributes by proposing a new prediction method, evaluating it against established techniques, highlighting deep learning's strengths and limitations, and demonstrating the potential for improved safety measures through machine learning. Ultimately, this work advances seismic monitoring methods for mining and oil exploration contexts.

show abstract

Section: Figure 1 Types Of Seismic Wavesmentioning

confidence: 99%

Seismic Waves Near Oil Reservoir Prediction Using Deep Learning

2023

HNSJ

View full text Add to dashboard Cite

show abstract

“…In seismic hazards analysis, they are a key tool for quantifying the ground motion of potential earthquakes (Cui et al, 2010). In oil and gas prospecting, they allow the seismic response of hydrocarbon reservoirs to be modelled (Lubrano-Lavadera et al, 2017;Siddiqui et al, 2017). In geophysical surveying, they show how the subsurface is illuminated by different survey designs (Xie et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

Deep learning for fast simulation of seismic waves in complex media

2020

View full text Add to dashboard Cite

Abstract. The simulation of seismic waves is a core task in many geophysical applications. Numerical methods such as finite difference (FD) modelling and spectral element methods (SEMs) are the most popular techniques for simulating seismic waves, but disadvantages such as their computational cost prohibit their use for many tasks. In this work, we investigate the potential of deep learning for aiding seismic simulation in the solid Earth sciences. We present two deep neural networks which are able to simulate the seismic response at multiple locations in horizontally layered and faulted 2-D acoustic media an order of magnitude faster than traditional finite difference modelling. The first network is able to simulate the seismic response in horizontally layered media and uses a WaveNet network architecture design. The second network is significantly more general than the first and is able to simulate the seismic response in faulted media with arbitrary layers, fault properties and an arbitrary location of the seismic source on the surface of the media, using a conditional autoencoder design. We test the sensitivity of the accuracy of both networks to different network hyperparameters and show that the WaveNet network can be retrained to carry out fast seismic inversion in the same media. We find that are there are challenges when extending our methods to more complex, elastic and 3-D Earth models; for example, the accuracy of both networks is reduced when they are tested on models outside of their training distribution. We discuss further research directions which could address these challenges and potentially yield useful tools for practical simulation tasks.

show abstract

“…In earthquake monitoring, they are a key tool for quantifying the ground motion of potential earthquakes [1]. In oil and gas prospecting, they are used to understand the seismic response of hydrocarbon reservoirs [2,3]. In geophysical surveying, they show how the subsurface is illuminated by different survey designs [4].…”

Section: Introductionmentioning

confidence: 99%

Fast approximate simulation of seismic waves with deep learning

Moseley¹,

Markham²,

Nissen‐Meyer³

2018

Preprint

View full text Add to dashboard Cite

We simulate the response of acoustic seismic waves in horizontally layered media using a deep neural network. In contrast to traditional finite-difference modelling techniques our network is able to directly approximate the recorded seismic response at multiple receiver locations in a single inference step, without needing to iteratively model the seismic wavefield through time. This results in an order of magnitude reduction in simulation time from the order of 1 s for FD modelling to the order of 0.1 s using our approach. Such a speed improvement could lead to real-time seismic simulation applications and benefit seismic inversion algorithms based on forward modelling, such as full waveform inversion. Our proof of concept deep neural network is trained using 50,000 synthetic examples of seismic waves propagating through different 2D horizontally layered velocity models. We discuss how our approach could be extended to arbitrary velocity models. Our deep neural network design is inspired by the WaveNet architecture used for speech synthesis. We also investigate using deep neural networks for simulating the full seismic wavefield and for carrying out seismic inversion directly.

show abstract

Seismic Modelling: 4D Capabilities for CO2 Injection

Cited by 4 publications

References 10 publications

Seismic Waves Near Oil Reservoir Prediction Using Deep Learning

Seismic Waves Near Oil Reservoir Prediction Using Deep Learning

Deep learning for fast simulation of seismic waves in complex media

Fast approximate simulation of seismic waves with deep learning

Contact Info

Product

Resources

About