Systems and methods for hybrid model hydraulic fracture pressure forecasting
A system for determining pressure in a hydraulic fracturing system for a well includes a processing module executing code and configured to receive a plurality of input parameters. The processing module can predict either a bottomhole pressure, based on statistical predictions and physics-based predictions, or a surface pressure based on the predicted bottomhole pressure.
1. A system for forecasting pressures in a hydraulic fracturing system for a well, the system comprising:
a processing module executing code and configured to:
receive specifications for performing a hydraulic fracturing operation on a well, the specifications forecasting an array of flow rate information for frac fluid over time;
submit the array of flow rate information to a model incorporating a statistically based multivariate relationship operating on the flowrate information to forecast pressure results including at least one of a bottomhole pressure time-based array and a surface pressure time based array,
each of the pressure results being separable into pressure loss subcomponents,
determine a system of multipliers for statistical normalization of the pressure loss subcomponents,
calculate normalized pressure loss subcomponents obtained by multiplying the pressure loss subcomponents by selected multipliers obtained from the system of multipliers, and
provide normalized pressure forecast data on a basis of the normalized pressure loss subcomponents;
monitor observed pressure from the well as the hydraulic fracturing operation is underway by comparing the normalized pressure forecast data over time to the observed pressure from the well; and
wherein the processing module is further configured to take responsive action when monitoring of the observed pressure indicates an impending screen-out, wherein such responsive action is selected from (1) adding a friction reducer to the frac fluid to increase flowrates while maintaining an operational pressure that does not exceed downhole pressure safety limits, (2) increasing flow rate by increasing pumping pressure while maintaining an operational pressure that does not exceed downhole pressure safety limits; or (3) reducing a bottomhole proppant concentration by mixing less proppant into the frac fluid.
2. The system of claim 1 wherein the array of flow rate information is associated with a second set of pressure forecast results obtained from a lumped 3D hydraulic fracturing model.
3. The system of claim 1 wherein the training data set incorporates historical data from at least one analogous hydraulic fracturing operation, the historical data including time-based arrays of at least two variables selected from the group consisting of historical flow rate (Q), historical surface pressure data (P surface ), historic bottomhole pressure data (P bottomhole ), entry pressure (P entry ), bottomhole proppant concentration (C BH ), net pressure (P net ), friction reducer concentration (FR), and fracture closure pressure (σ closure ).
4. The system of claim 3 wherein the time-based arrays include at least three of the variables.
5. The system of claim 3 wherein the time-based arrays include at least four of the variables.
6. The system of claim 3 wherein the time-based arrays include at least five of the variables.
7. The system of claim 3 wherein the time-based arrays include at least six of the variables.
8. The system of claim 3 wherein the training data contains historical information from diagnostic fracture injection tests, and the statistical model relates the historical information from diagnostic fracture injection tests to new data obtained from a diagnostic fracture injection test performed on the well.
9. The system of claim 3 wherein the training data contains historical information from a stepdown test performed on the well, and the statistical model relates historical information from diagnostic fracture injection tests to new data obtained from the stepdown test.
10. The system of claim 3 wherein the training data set includes historical data from a prior stage of hydraulic fracturing performed on the well.
11. The system of claim 3 wherein the training data set includes historical data from the same stage of the hydraulic fracturing operation as the hydraulic fracturing operation of the same stage remains underway.
12. The system of claim 1 wherein the system of multipliers includes at least one relationship for determining net pressure multipliers (M net ) for mitigating error from statistical modeling of net pressure (P net ) based upon data obtained from stepdown tests performed on the plurality of wells.
13. The system of claim 1 wherein the system of multipliers includes at least one relationship for mitigating error in pressure loss from wellbore friction (P wellfriction,Q,C ) as a function of flowrate (Q).
14. The system of claim 1 wherein the system of multipliers includes at least one relationship for mitigating error in pressure loss from wellbore friction (P wellfriction,Q,C ) as a function of bottomhole proppant concentration (C BH ).
15. The system of claim 1 wherein the responsive action is to initiate an alarm to alert a frac operator when a risk of screen-out is elevated.
16. The system of claim 1 wherein the code is further configured to interact with a user through use of a graphical user interface to facilitate the responsive action.
17. The system of claim 1 wherein the responsive action is determined as a result of threshold analysis.
18. The system of claim 17 wherein the threshold analysis utilizes a probabilistic threshold on a plot of actual pressure (P actual ) versus forecast pressure (P forecast ).
19. The system of claim 18 wherein the probabilistic threshold is based upon a Gaussian distribution utilizing as a mode of the Gaussian distribution a least squares fit of a dataset comprised of the (P actual , P forecast ) values.
20. The system of claim 17 wherein the threshold analysis is based upon a contemporaneous comparison of the observed pressure and the normalized pressure forecast data.