Method for determining properties of a formation
A method for determining a property of a formation, including the steps: drilling a well in the formation, collecting drill cuttings from the well, taking a digital image of each drill cutting, entering each digital image to a trained first model that outputs a predicted lithology class of each drill cutting from each digital image, taking a random number of X-ray diffraction (XRD) images of the drill cuttings, while at least one XRD image is selected from each lithology class, entering each XRD image and the corresponding digital image into a trained second model that predicts a property of the drill cuttings, and determining the property of the formation by determining the properties of the drill cuttings as a function of the depth of the drill cuttings.
1. A method for determining a property of a formation, comprising the steps:
drilling a well in the formation,
collecting drill cuttings from the well,
taking a digital image of each drill cutting,
entering each digital image to a trained first model that outputs a predicted lithology class of each drill cutting from each digital image,
taking a random number of X-ray diffraction (XRD) images of the drill cuttings, wherein at least one XRD image is selected from each predicted lithology class,
entering each XRD image and the corresponding digital image into a trained second model that predicts a property of the drill cuttings, and
determining the property of the formation by determining the properties of the drill cuttings as a function of the depth of the drill cuttings.
2. The method of claim 1 , wherein the property of the formation comprises mineral composition, grain size, porosity, and rock type.
3. The method of claim 1 , wherein the first model is verified by constant monitoring of the predicted lithology classes.
4. The method of claim 1 , wherein the second model is verified by periodically monitoring the properties of the drill cuttings.
5. The method of claim 1 , wherein the properties of the drill cuttings are monitored by comparing the predicted class with the actual class of the properties of the drill cuttings.
6. The method of claim 1 , wherein the first model comprises an artificial intelligence (AI) model.
7. The method of claim 1 , wherein the first model comprises a machine learning model.
8. The method of claim 1 , wherein the first model comprises a deep learning model.
9. The method of claim 1 , wherein the second model comprises an artificial intelligence (AI) model.
10. The method of claim 1 , wherein the second model comprises a machine learning model.
11. The method of claim 1 , wherein the second model comprises a deep learning model.
12. The method of claim 1 , wherein the XRD images are calibrated to industry standard.
13. The method of claim 1 , wherein the digital images are in the following data format: gif, jpeg, tiff, png, eps, or raw.
14. The method of claim 1 , wherein the XRD images are in the following data format: gif, jpeg, tiff, png, eps, or raw.
15. The method of claim 1 , wherein the first model is embedded in a first software.
16. The method of claim 1 , wherein the second model is embedded in a second software.
17. The method of claim 1 , wherein the first and second model are embedded in a single software.
18. The method of claim 1 , wherein the drill cuttings are collected at every 3 meter (10 feet) of the wellbore.
19. The method of claim 1 , wherein digital images of drill cuttings with low accuracy of the predicted lithology class are used to create a new lithology class.
20. The method of claim 1 , wherein the digital images comprise high-resolution images with at least 300 ppi.