Multimodal well log interpretation using deep learning models

1. A computer-implemented method using cognitive analysis for well log interpretation, comprising:

receiving, by one or more processors, a plurality of well logs corresponding to one or more wells within an oilfield;

extracting, by the one or more processors, well log data from the plurality of well logs, the extracted well log data comprising geometric features including characteristics of each log curve displayed on one or more tracks of the plurality of well logs, specific values associated with petrophysical properties measured by each of the plurality of well logs, and well attributes associated with well characteristics and operational parameters displayed on each of the plurality of well logs;

embedding, by the one or more processors, the geometric features, the specific values and the well attributes;

based on the embedded geometric features, specific values and well attributes, generating, by the one or more processors, intermediate vectors;

assigning, by the one or more processors, a weight to each intermediate vector according to a relevance for identifying petrophysical properties and stratigraphic layers;

determining, by the one or more processors, a correlation between the intermediate vectors and a target reservoir layer;

based on a distance between an intermediate vector associated with the target reservoir layer and a target vector being less than a predefined value, assigning the geometric features, the specific values and the well attributes to the target reservoir layer; and

responsive to the intermediate vector being within a confidence interval, selecting, by the one or more processors, the intermediate vector associated with the target reservoir layer as a fusion vector for determining hydrocarbon bearing zones.

2. The method of claim 1 , wherein the characteristics of each log curve displayed in the one or more tracks of the plurality of well logs included in the geometric features are selected from the group consisting of a number of log curves available within the each of the plurality of well logs, a scale, a distance between different log curves, an area of the log curves within specific segments, a gradient of each log curve, and a variance of each log curve.

3. The method of claim 1 , wherein the specific values comprise a rock type, a grain size and a bed thickness.

4. The method of claim 1 , wherein the well attributes associated with the well characteristics and the operational parameters are selected from the group consisting of a well name and location, a well type and head size, a logging company, a name of an energy company, a type of logs recorded on a run, a date of a logging operation, a ground elevation and Kelly bushing elevation, a drill-bit size, a drilling mud type, a total depth according to driller and logging engineer, a bottom-hole temperature, an interval that was logged, and drilling mud characteristics.

5. The method of claim 1 , wherein the plurality of well logs are received using at least one of a .LAS file and an image file.

6. The method of claim 1 , wherein the plurality of well logs includes one or more of a gamma-ray log, a neutron-porosity log, a resistivity log, and a spontaneous potential log.

7. The method of claim 1 , wherein identifying the reservoir layers based on the fusion vector is based on a convolutional neural networks approach.

8. A computer system using cognitive analysis for well log interpretation, comprising:

one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage devices, and program instructions stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:

receiving, by one or more processors, a plurality of well logs corresponding to one or more wells within an oilfield;

extracting, by the one or more processors, well log data from the plurality of well logs, the extracted well log data comprising geometric features including characteristics of each log curve displayed on one or more tracks of the plurality of well logs, specific values associated with petrophysical properties measured by each of the plurality of well logs, and well attributes associated with well characteristics and operational parameters displayed on each of the plurality of well logs;

embedding, by the one or more processors, the geometric features, the specific values and the well attributes;

based on the embedded geometric features, specific values and well attributes, generating, by the one or more processors, intermediate vectors;

assigning, by the one or more processors, a weight to each intermediate vector according to a relevance for identifying petrophysical properties and stratigraphic layers;

determining, by the one or more processors, a correlation between the intermediate vectors and a target reservoir layer;

based on a distance between an intermediate vector associated with the target reservoir layer and a target vector being less than a predefined value, assigning the geometric features, the specific values and the well attributes to the target reservoir layer; and

responsive to the intermediate vector being within a confidence interval, selecting, by the one or more processors, the intermediate vector associated with the target reservoir layer as a fusion vector for determining hydrocarbon bearing zones.

9. The computer system of claim 8 , wherein the characteristics of each log curve displayed in the one or more tracks of the plurality of well logs included in the geometric features are selected from the group consisting of a number of log curves available within the each of the plurality of well logs, a scale, a distance between different log curves, an area of the log curves within specific segments, a gradient of each log curve, and a variance of each log curve.

10. The computer system of claim 8 , wherein the specific values comprise a rock type, a grain size and a bed thickness.

11. The computer system of claim 8 , wherein the well attributes associated with the well characteristics and the operational parameters are selected from the group consisting of a well name and location, a well type and head size, a logging company, a name of an energy company, a type of logs recorded on a run, a date of a logging operation, a ground elevation and Kelly bushing elevation, a drill-bit size, a drilling mud type, a total depth according to driller and logging engineer, a bottom-hole temperature, an interval that was logged, and drilling mud characteristics.

12. The computer system of claim 8 , wherein the plurality of well logs are received using at least one of a .LAS file and an image file.

13. The computer system of claim 8 , wherein the plurality of well logs includes one or more of a gamma-ray log, a neutron-porosity log, a resistivity log, and a spontaneous potential log.

14. The computer system of claim 8 , wherein identifying the reservoir layers based on the fusion vector is based on a convolutional neural networks approach.

15. A computer program product using cognitive analysis for well log interpretation, comprising:

a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a device to cause the device to perform a method comprising:

receiving, by one or more processors, a plurality of well logs corresponding to one or more wells within an oilfield;

extracting, by the one or more processors, well log data from the plurality of well logs, the extracted well log data comprising geometric features including characteristics of each log curve displayed on one or more tracks of the plurality of well logs, specific values associated with petrophysical properties measured by each of the plurality of well logs, and well attributes associated with well characteristics and operational parameters displayed on each of the plurality of well logs;

embedding, by the one or more processors, the geometric features, the specific values and the well attributes;

based on the embedded geometric features, specific values and well attributes, generating, by the one or more processors, intermediate vectors;

assigning, by the one or more processors, a weight to each intermediate vector according to a relevance for identifying petrophysical properties and stratigraphic layers;

determining, by the one or more processors, a correlation between the intermediate vectors and a target reservoir layer;

based on a distance between an intermediate vector associated with the target reservoir layer and a target vector being less than a predefined value, assigning the geometric features, the specific values and the well attributes to the target reservoir layer; and

responsive to the intermediate vector being within a confidence interval, selecting, by the one or more processors, the intermediate vector associated with the target reservoir layer as a fusion vector for determining hydrocarbon bearing zones.

16. The computer program product of claim 15 , wherein the characteristics of each log curve displayed in the one or more tracks of the plurality of well logs included in the geometric features are selected from the group consisting of a number of log curves available within the each of the plurality of well logs, a scale, a distance between different log curves, an area of the log curves within specific segments, a gradient of each log curve, and a variance of each log curve.

17. The computer program product of claim 15 , wherein the specific values comprise a rock type, a grain size and a bed thickness.

18. The computer program product of claim 15 , wherein the well attributes associated with the well characteristics and the operational parameters are selected from the group consisting of a well name and location, a well type and head size, a logging company, a name of an energy company, a type of logs recorded on a run, a date of a logging operation, a ground elevation and Kelly bushing elevation, a drill-bit size, a drilling mud type, a total depth according to driller and logging engineer, a bottom-hole temperature, an interval that was logged, and drilling mud characteristics.

19. The computer program product of claim 15 , wherein the plurality of well logs are received using at least one of a .LAS file and an image file.

20. The computer program product of claim 15 , wherein the plurality of well logs includes one or more of a gamma-ray log, a neutron-porosity log, a resistivity log, and a spontaneous potential log.