Predicting drilling tool failure

1. A method for predicting drilling tool failure, which comprises:

computing axial, torsional, and lateral jerk values for a drilling tool using accelerometer data and a computer processor;

computing axial, torsional, and lateral inverse jerk values for the drilling tool using a respective jerk value and the computer processor;

plotting the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values on a graph relative to a time;

determining failure threshold limits for the drilling tool based on the graph; plotting the failure threshold limits and warning thresholds for the drilling tool on the graph relative to a time;

predicting drilling tool failure using the graph with the failure threshold limits and warning thresholds;

determining historical drilling tool failure data by running a data driven model simulation using historical data from prior drilling operations;

determining drilling tool failure pattern trends data using the historical drilling tool failure data;

plotting the drilling tool failure pattern trends data on the graph with the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values relative to a time;

determining new failure threshold limits for the drilling tool based on the graph with the drilling tool failure pattern trends data;

plotting the new failure threshold limits and new warning thresholds for the drilling tool on the graph with the drilling tool failure pattern trends data relative to a time; and

predicting drilling tool failure using the graph with the new failure threshold limits and the new warning thresholds.

2. The method of claim 1 , further comprising adjusting drilling operations based on the predicted drilling tool failure.

3. The method of claim 1 , wherein the warning thresholds and the new warning thresholds are based on operational cycles and a performance baseline for the drilling tool.

4. The method of claim 3 , wherein the operational cycles and the performance baseline for the drilling tool are based on at least one of drilling tool parameters of an operating environment; laboratory checks; quality control reports; predeployment test statistics; design ratings; and manufacture's specifications.

5. The method of claim 4 , wherein the parameters of the operating environment for the drilling tool are based on subsurface conditions and a predetermined drilling program.

6. The method of claim 1 , wherein the historical data from prior drilling operations includes at least one of prior drilling tool failure logs, drilling tool operating cycles before failure, formation properties, subsurface conditions and historical accelerometer data, each from previously drilled wells in geologically similar areas.

7. The method of claim 1 , wherein the data driven model simulation is run using a generalized regression neural networks data driven model.

8. A non-transitory computer-readable storage medium storing computer executable instructions for predicting drilling tool failure, the instructions being executable by one or more processors to implement:

computing axial, torsional, and lateral jerk values for a drilling tool using accelerometer data;

computing axial, torsional, and lateral inverse jerk values for the drilling tool using a respective jerk value;

plotting the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values on a graph relative to a time (t);

determining failure threshold limits for the drilling tool based on the graph; plotting the failure threshold limits and warning thresholds for the drilling tool on the graph relative to a time;

predicting drilling tool failure using the graph with the failure threshold limits and warning thresholds;

determining historical drilling tool failure data by running a data driven model simulation using historical data from prior drilling operations;

determining drilling tool failure pattern trends data using the historical drilling tool failure data;

plotting the drilling tool failure pattern trends data on the graph with the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values relative to a time;

determining new failure threshold limits for the drilling tool based on the graph with the drilling tool failure pattern trends data;

plotting the new failure threshold limits and new warning thresholds for the drilling tool on the graph with the drilling tool failure pattern trends data relative to a time; and

predicting drilling tool failure using the graph with the new failure threshold limits and the new warning thresholds.

9. The non-transitory computer-readable storage medium of claim 8 , further comprising adjusting drilling operations based on the predicted drilling tool failure.

10. The non-transitory computer-readable storage medium of claim 8 , wherein the warning thresholds and the new warning thresholds are based on operational cycles and a performance baseline for the drilling tool.

11. The non-transitory computer-readable storage medium of claim 10 , wherein the operations cycles and the performance baseline for the drilling tool are based on at least one of drilling tool parameters of an operating environment; laboratory checks; quality control reports; predeployment test statistics; design ratings; and manufacturer's specifications.

12. The non-transitory computer-readable storage medium of claim 11 , wherein the parameters of the operating environment for the drilling tool are based on subsurface conditions and a predetermined drilling program.

13. The non-transitory computer-readable storage medium of claim 8 , wherein the historical data from prior drilling operations includes at least one of prior drilling tool failure logs, drilling tool operating cycles before failure, formation properties, subsurface conditions and historical accelerometer data, each from previously drilled wells in geologically similar areas.

14. The non-transitory computer-readable storage medium of claim 8 , wherein the data driven model simulation is run using a generalized regression neural networks data driven model.

15. A non-transitory computer-readable storage medium storing computer executable instructions for predicting drilling tool failure, the instructions being executable by one or more processors to implement:

computing axial, torsional, and lateral jerk values for a drilling tool using accelerometer data;

computing axial, torsional, and lateral inverse jerk values for the drilling tool using a respective jerk value;

plotting the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values on a graph relative to a time;

determining failure threshold limits for the drilling tool based on the graphs; plotting the failure threshold limits and warning thresholds for the drilling tool on the graph relative to a time;

adjusting drilling operations based on the failure threshold limits and warning thresholds;

determining historical drilling tool failure data by running a data driven model simulation using historical data from prior drilling operations;

determining drilling tool failure pattern trends data using the historical drilling tool failure data;

plotting the drilling tool failure pattern trends data on the graph with the axial, torsional, and lateral jerk values and the axial, torsional, and lateral inverse jerk values relative to a time;

determining new failure threshold limits for the drilling tool based on the graph with the drilling tool failure pattern trends data;

plotting the new failure threshold limits and new warning thresholds for the drilling tool on the graph with the drilling tool failure pattern trends data relative to a time; and

predicting drilling tool failure wising the graph with the new failure threshold limits and the new warning thresholds.

16. The method of claim 1 , wherein the accelerometer data includes at least one of real-time data and historical data.

17. The non-transitory computer-readable storage medium of claim 8 , wherein the accelerometer data includes at least one of real-time data and historical data.