Predicting formation frackability using surface logging data

1 . A computer-implemented method, comprising:

accessing surface logging data of a well being drilled for fracking, the surface logging data being converted into a table comprising rows and columns;

performing quality control (QC) and quality assurance (QA) on the surface logging data;

performing drift analysis on the surface logging data to ensure probability distribution of the surface logging data to be within a predetermined value;

verifying a terminology consistency across the rows and columns of the table to identify inconsistent values associated with drilling parameters and mud logs;

filtering the inconsistent values from the table to generate prepared surface logging data;

generating, based on processing the prepared surface logging data using a predictive model, a visualization formatted as a depth chart and comprising information provided for different drilling depths displayed as a plurality of stages coded to indicate a probability of frackability in a track with other log tracks;

determining frackability probabilities for fracking intervals relative to the different drilling depths; and

updating wellbore operations to adjust settings of a downhole equipment according to the fracking intervals.

2 . The computer-implemented method of claim 1 , wherein the surface logging data comprises the drilling parameters, mud log lithology, and chromatography gas analysis data.

3 . The computer-implemented method of claim 1 , wherein filtering the inconsistent values comprises checking data quality row-by-row and column-by-column, filtering out rows with one or both of missing and inconsistent values, and profiling data against measured depths to show missing values.

4 . The computer-implemented method of claim 1 , wherein performing drift analysis on the surface logging data comprises applying Wasserstein distance and Jensen-Shannon metrics to avoid incompatible datasets.

5 . The computer-implemented method of claim 1 , wherein the visualization comprises color coded intervals for different ranges of frackability.

6 . The computer-implemented method of claim 1 , further comprising:

generating plots of predicative model accuracies based on blind testing of randomly selected data.

7 . The computer-implemented method of claim 6 , wherein the plots display model accuracy as a function of a number of resampling iterations in which a random dataset sample serves as a blind data set to evaluate the model accuracy.

8 . A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

accessing surface logging data of a well being drilled for fracking, the surface logging data being converted into a table comprising rows and columns;

performing quality control (QC) and quality assurance (QA) on the surface logging data;

performing drift analysis on the surface logging data to ensure probability distribution of the surface logging data to be within a predetermined value;

verifying a terminology consistency across the rows and columns of the table to identify inconsistent values associated with drilling parameters and mud logs;

filtering the inconsistent values from the table to generate prepared surface logging data;

generating, based on processing the prepared surface logging data using a predictive model, a visualization formatted as a depth chart and comprising information provided for different drilling depths displayed as a plurality of stages coded to indicate a probability of frackability in a track with other log tracks;

determining frackability probabilities for fracking intervals relative to the different drilling depths; and

updating wellbore operations to adjust settings of a downhole equipment according to the fracking intervals.

9 . The non-transitory, computer-readable medium of claim 8 , wherein the surface logging data comprises the drilling parameters, mud log lithology, and chromatography gas analysis data.

10 . The non-transitory, computer-readable medium of claim 8 , wherein filtering the inconsistent values comprises checking data quality row-by-row and column-by-column, filtering out rows with one or both of missing and inconsistent values, and profiling data against measured depths to show missing values.

11 . The non-transitory, computer-readable medium of claim 8 , wherein performing drift analysis on the surface logging data comprises applying Wasserstein distance and Jensen-Shannon metrics to avoid incompatible datasets.

12 . The non-transitory, computer-readable medium of claim 8 , wherein the visualization comprises color coded intervals for different ranges of frackability.

13 . The non-transitory, computer-readable medium of claim 8 , the operations further comprising:

generating plots of predicative model accuracies based on blind testing of randomly selected data.

14 . The non-transitory, computer-readable medium of claim 13 , wherein the plots display model accuracy as a function of a number of resampling iterations in which a random dataset sample serves as a blind data set to evaluate the model accuracy.

15 . A computer-implemented system, comprising:

one or more processors; and

a non-transitory computer-readable storage medium coupled to the one or more processors and storing programming instructions for execution by the one or more processors, the programming instructions instructing the one or more processors to perform operations comprising:

accessing surface logging data of a well being drilled for fracking, the surface logging data being converted into a table comprising rows and columns;

performing quality control (QC) and quality assurance (QA) on the surface logging data;

performing drift analysis on the surface logging data to ensure probability distribution of the surface logging data to be within a predetermined value;

verifying a terminology consistency across the rows and columns of the table to identify inconsistent values associated with drilling parameters and mud logs;

filtering the inconsistent values from the table to generate prepared surface logging data;

generating, based on processing the prepared surface logging data using a predictive model, a visualization formatted as a depth chart and comprising information provided for different drilling depths displayed as a plurality of stages coded to indicate a probability of frackability in a track with other log tracks;

determining frackability probabilities for fracking intervals relative to the different drilling depths; and

updating wellbore operations to adjust settings of a downhole equipment according to the fracking intervals.

16 . The computer-implemented system of claim 15 , wherein the surface logging data comprises the drilling parameters, mud log lithology, and chromatography gas analysis data.

17 . The computer-implemented system of claim 15 , wherein filtering the inconsistent values comprises checking data quality row-by-row and column-by-column, filtering out rows with one or both of missing and inconsistent values, and profiling data against measured depths to show missing values.

18 . The computer-implemented system of claim 15 , wherein performing drift analysis on the surface logging data comprises applying Wasserstein distance and Jensen-Shannon metrics to avoid incompatible datasets.

19 . The computer-implemented system of claim 15 , wherein the visualization comprises color coded intervals for different ranges of frackability.

20 . The computer-implemented system of claim 15 , the operations further comprising:

generating plots of predicative model accuracies based on blind testing of randomly selected data.