Automatic, geologic layer-constrained well-seismic tie through blocked dynamic warping

Abstract

Well-log-to-seismic tying is a key step in many interpretation workflows for oil and gas exploration. Synthetic seismic traces from the wells are often manually tied to seismic data; this process can be very time consuming and, in some cases, inaccurate. Automatic methods, such as dynamic time warping (DTW), can match synthetic traces to seismic data. Although these methods are extremely fast, they tend to create interval velocities that are not geologically realistic. We have described the modification of DTW to create a blocked dynamic warping (BDW) method. BDW generates an automatic, optimal well tie that honors geologically consistent velocity constraints. Consequently, it results in updated velocities that are more realistic than other methods. BDW constrains the updated velocity to be constant or linearly variable inside each geologic layer. With an optimal correlation between synthetic seismograms and surface seismic data, this algorithm returns an automatically updated time-depth curve and an updated interval velocity model that still retains the original geologic velocity boundaries. In other words, the algorithm finds the optimal solution for tying the synthetic to the seismic data while restricting the interval velocity changes to coincide with the initial input blocking. We have determined the application of the BDW technique on a synthetic data example and field data set.

A Method for Defining Sedimentary Characteristics and Distributions and Its Application in Qinnan Depression, Bohai Bay Basin

“…Many methods have been used to define sedimentary characteristics and distributions for reservoir exploration, which can be roughly divided into five categories: (1) geological analysis based on sedimentology, sequence stratigraphy and seismic sedimentology, which includes core or outcrop observation, a comprehensive analysis of seismic facies and log facies and an analysis of the source-to-sink system centering on lithology and geomorphology [51][52][53]; (2) geophysical analysis, which contains seismic forward modeling, seismic inversion and the reprocessing of seismic data [11]; (3) geochemical analysis including elemental analysis and organic geochemical analysis [47,49]; (4) hybrid analysis, which assembles results derived from different methods to acquire a comprehensive result [8,11,16,53]; and (5) the artificial intelligent method, which is used to conduct a comprehensive data analysis based on a large database [11]. Although the methods above have each their own advantages, they rarely have a closed-loop research process to ensure the correctness and uniqueness of the analysis results.…”

“…Representing the lithofacies' transition surfaces, the peak reflection and trough reflection are, respectively, regarded as the interface from low to high impedance and the interface from high to low impedance [53]. Thus, it is most appropriate to leverage the maximum-peak amplitude to distinguish sand-rich deposition from mud-rich deposition, and then to depict the plane distribution of the sand-rich deposition.…”

70 years of machine learning in geoscience in review

“…The “hidden” part of Markov models enables the model to assume influences on the predictions that are not directly represented in the input data. The K-nearest neighbors method has been used for well-log analysis ( Caté, Perozzi, Gloaguen, & Blouin, 2017 ; Saporetti et al, 2018 ), seismic well ties ( Wang, Lomask, & Segovia, 2017 ) combined with dynamic time warping and fault extraction in seismic interpretation ( Hale, 2013 ), which is highly dependent on choosing the right hyperparameter k. The unsupervised k-NN equivalent, k-means has been applied to seismic interpretation ( Di, Shafiq, & AlRegib, 2017 a ), ground motion model validation ( Khoshnevis & Taborda, 2018 ), and seismic velocity picking ( Wei, Yonglin, Qingcai, Jiaqiang, et al, 2018 ). These are very simple machine learning models that are useful for baseline models.…”

70 years of machine learning in geoscience in review

“…The "hidden" part of Markov models enables the model to assume influences on the predictions that are not directly represented in the input data. The K-nearest neighbours method has been used for well-log analysis , Saporetti et al, 2018, seismic well ties [Wang et al, 2017b] combined with dynamic time warping and fault extraction in seismic interpretation [Hale, 2013], which is highly dependent on choosing the right hyperparameter k. The unsupervised k-means equivalent has been applied to seismic interpretation [Di et al, 2017a], ground motion model validation [Khoshnevis and Taborda, 2018], and seismic velocity picking [Wei et al, 2018]. These are very simple machine learning models that are useful for baseline models.…”