Deep Seismic2Well Tie: A physics-guided CNN approach to a classic geophysical workflow

Thanoon, David; Vamaraju, Janaki; Yang, Heng; Wei, Katy; Aussavy, Antoine Vial; Chen, Jie

doi:10.1190/segam2021-3587358.1

Cited by 3 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Nivlet et al (2020) implemented both the supervised LSTM and TCN for automatically converting sonic log from depth to time. For addressing the unavailability of sufficient bias-free human-interpreted labels, Thanoon et al (2021) adopted the physics-guided convolutional neural network (CNN) (Biswas et al, 2019) for estimating the TDRs. Wu et al (2022) proposed using a CNN for segment-based TDR stretching and squeezing, which requires simulating all possible velocity models at a target well to generate training data and repeating the workflow from one well to another.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automating seismic‐well tie via self‐supervised learning

Abubakar

2023

Geophysical Prospecting

View full text Add to dashboard Cite

As an essential process in subsurface interpretation, seismic-well tie aims at calibrating the well measurements in depth domain with the seismic records in time domain for reliable reservoir mapping and modelling. Such a process usually requires massive manual efforts, primarily extracting source wavelets to synthesize seismograms from well measurements and stretching/squeezing synthetic seismograms to match actual seismic signals, and thus becomes time-consuming and labour intensive with the amount of wells expanding in a field. This paper formulates the seismic-well tie problem from the perspective of computer vision and, considering the lack of manually prepared training labels in most of real projects, proposes automating it by a self-supervised workflow of two key components. The first component is to extract time-variant wavelets in the target interval using a dual-task autoencoder, which is optimized by maximizing the spectrum similarity between the extracted wavelet and the actual seismic. The second component is to estimate the corresponding time-shift between a pair of the synthetic seismogram and the actual seismic using a flow net, which is trained by a pseudo dataset derived from the actual well measurements. More specifically, the data generation consists with defining one-dimensional time-shift curves to warp the original less accurate time-depth relationships and then synthesizing the corresponding seismograms based on the convolutional model at all available wells. When turning to the stage of inference at a target well, the proposed workflow automatically extracts a representative wavelet, generates the corresponding synthetic seismogram from the original timedepth relationship and estimates the time-shift curve necessary for revising the original time-depth relationship that would lead to an optimal tying with the actual seismic at the well. The proposed workflow is tested on two field datasets, and both results demonstrate improved tying over traditional methods such as statistical wavelet extraction and dynamic time warping.

show abstract

Section: Introductionmentioning

confidence: 99%

“…For addressing the unavailability of sufficient bias‐free human‐interpreted labels, Thanoon et al. (2021) adopted the physics‐guided convolutional neural network (CNN) (Biswas et al., 2019) for estimating the TDRs. Wu et al.…”

Section: Introductionmentioning

confidence: 99%

Automating seismic‐well tie via self‐supervised learning

Abubakar

2023

Geophysical Prospecting

View full text Add to dashboard Cite

show abstract

Full-Stack Machine Learning Development Framework for Energy Industry Applications

Chen

Yao

Devarakota

et al. 2022

Day 3 Wed, November 02, 2022

View full text Add to dashboard Cite

Machine Learning (ML) has proved successful in various applications and delivered tremendous value across numerous domains. ML turns data into knowledge and intelligence, that can be used to make the right business decisions. The application of ML in the energy industry is increasing rapidly. This includes but is not limited to manufacturing, refining, energy distribution, and other related domains. Due to the unique and diverse domain requirements, various ML and AI solutions in the energy industry must be extensively customized. These specific requirements usually lead to different approaches to solutions and further introduce difficulties in developing, deploying, and scaling. These difficulties lead to higher costs and longer cycle time in productization. To reduce cost and save time, there is a growing business need to develop, deploy, and scale these energy industry ML applications in an agile way. To achieve this goal, we developed a full-stack ML development framework, which makes it easier to develop and deploy (energy industry) ML solutions with high efficiency, reproducibility, and scalability. It has been demonstrated that projects using this full-stack ML development framework successfully saved both turnaround time and costs.

show abstract

Self-Supervised Learning for Automated Seismic Wavelet Extraction

Abubakar

2022

Day 2 Tue, November 01, 2022

View full text Add to dashboard Cite

Seismic wavelet extraction is a fundamental but crucial component in seismic data analysis, which aims at estimating the source wavelet for better decoding the subsurface reflectivities from seismic signals and calibrating the measurements between logging and seismic. Due to the rock heterogeneity, however, seismic energy attenuates during its propagation, and correspondingly the seismic wavelet is considered varying from shallow to deep as well as from near to far offsets, causing the wavelet time-variant. While machine learning (ML) appears feasible for assisting the challenge, without a comprehensive understanding of the seismic propagation, representative training labels cannot be prepared, and supervised learning appears less applicable. This study proposes implementing self-supervised learning into time-variant seismic wavelet extraction. Specifically, the proposed network is in the architecture of a dual-task auto-encoder (DTAE). Starting from 1-D seismic amplitude, the DTAE first uses an encoder to extract a set of features at multiple scales, which are then split into two flows, with one through a decoder that aims at reconstructing the input 1D seismic signal and the other through a few convolutional layers that aim at matching the spectrum between seismic trace and extracted wavelet. Correspondingly, the objective function of training such DTAE consists of two parts, the mean-square-error of the amplitude reconstruction and the mean-square-error of the spectrum matching. The proposed workflow is tested on various field datasets, and compared to the statistical wavelets, the extracted wavelets are of better match with the spectrum variation of seismic from shallow to deep and from near offset to far offset.

show abstract

Deep Seismic2Well Tie: A physics-guided CNN approach to a classic geophysical workflow

Cited by 3 publications

References 8 publications

Automating seismic‐well tie via self‐supervised learning

Automating seismic‐well tie via self‐supervised learning

Full-Stack Machine Learning Development Framework for Energy Industry Applications

Self-Supervised Learning for Automated Seismic Wavelet Extraction

Contact Info

Product

Resources

About