Apparatus and method for fracking optimization

1 . An apparatus for fracking optimization, wherein the apparatus comprises:

at least a processor and a memory communicatively connected to the processor, the memory containing instructions configuring the processor to:

receive a plurality of reservoir datums from a sensor suite, wherein the plurality of reservoir datums comprises:

at least one surface reservoir datum detected using a first sensor of the sensor suite; and

at least one downhole reservoir datum detected using a second sensor of the sensor suite;

generate production training data as a function of the plurality of reservoir datums, wherein generating the production training data comprises:

processing the at least one downhole reservoir datum into a standardized data format using a data conversion module;

obtaining at least one fracking simulation model output, wherein the fracking simulation model is configured to calculate at least one of: rock deformation, reservoir pressure, fracture propagation, fluid flow in a fracture; and

correlating the plurality of reservoir datums with the at least one fracking simulation model output to generate the production training data;

 train a fracking optimization machine learning model using the pro duction training data, wherein the fracking optimization machine learning model is configured to predict the at least one fracking simulation model output without requiring execution of the fracking simulation model;

 receive a surface target reservoir datum from a sensor suite; and

 determine a fracking production parameter in real-time using the surface

target reservoir datum and the trained fracking optimization machine learning model without requiring execution of the fracking simulation model.

2 . The apparatus of claim 1 , wherein the fracking optimization machine learning model is configured to predict at least one of: fracture height, fracture width, fracture length, and fracture conductivity as the at least one fracking simulation model output.

3 . The apparatus of claim 1 , wherein the memory further contains instructions configuring the processor to:

predict, using the fracking optimization machine learning model, a likelihood of fracturing into a nearby wellbore based on the surface target reservoir datum.

4 . The apparatus of claim 1 , wherein the memory further contains instructions configuring the processor to:

generate a well spacing plan based on the fracking production parameter determined using the fracking optimization machine learning model.

5 . The apparatus of claim 1 , wherein the fracking optimization machine learning model is trained to map between the plurality of reservoir datums as inputs and the at least one fracking simulation model output as outputs, thereby eliminating a need for executing the fracking simulation model during real-time operations.

6 . The apparatus of claim 1 , wherein the fracking production parameter comprises at least one of: optimal drilling depth, optimal well location, optimal proximity of fractures, optimal quantity of fractures, optimal fracking fluid pressure, and optimal fracking fluid flow rates.

7 . The apparatus of claim 1 , wherein the fracking simulation model is configured to calculate fracture propagation using at least one fracture mechanics criterion, wherein the at least one fracture mechanics criterion comprises an evaluation of a plurality of reservoir datums.

8 . The apparatus of claim 1 , wherein the fracking simulation model is configured to simulate fluid flow in a fracture by implementing a fluid dynamics model that relates fluid parameters and downhole conditions to pressure gradient along the fracture, and wherein the fracking optimization machine learning model is trained to predict this fluid flow without executing the fracking simulation model.

9 . The apparatus of claim 1 , wherein the fracking simulation model is configured to calculate rock deformation as a function of reservoir datum using a rock behavior model, and wherein the fracking optimization machine learning model is trained to predict this rock deformation without executing the fracking simulation model.

10 . A method for fracking optimization, comprising:

receiving, by at least a processor, a plurality of reservoir datums from a sensor suite, wherein the plurality of reservoir datums comprises:

at least one surface reservoir datum detected using a first sensor of the sensor suite; and

at least one downhole reservoir datum detected using a second sensor of the sensor suite;

generating, by the at least a processor, production training data as a function of the plurality of reservoir datums, wherein generating the production training data comprises:

processing the at least one downhole reservoir datum into a standardized data format using a data conversion module;

obtaining at least one fracking simulation model output, wherein the fracking simulation model is configured to calculate at least one of: rock deformation, reservoir pressure, fracture propagation, fluid flow in a fracture; and

correlating the plurality of reservoir datums with the at least one fracking simulation model output to generate the production training data;

training, by the at least a processor, a fracking optimization machine learning model using the production training data, wherein the fracking optimization machine learning model is configured to predict the at least one fracking simulation model output without requiring execution of the fracking simulation model;

receiving, by the at least a processor, a surface target reservoir datum from a sensor suite; and

determining, by the at least a processor, a fracking production parameter in real-time using the surface target reservoir datum and the trained fracking optimization machine learning model without requiring execution of the fracking simulation model.

11 . The method of claim 10 , wherein the fracking optimization machine learning model is configured to predict at least one of: fracture height, fracture width, fracture length, and fracture conductivity as the at least one fracking simulation model output.

12 . The method of claim 10 , further comprising:

predicting, by the at least a processor using the fracking optimization machine learning model, a likelihood of fracturing into a nearby wellbore based on the surface target reservoir datum.

13 . The method of claim 10 , further comprising:

generating, by the at least a processor, a well spacing plan based on the fracking production parameter determined using the fracking optimization machine learning model.

14 . The method of claim 10 , wherein the fracking optimization machine learning model is trained to map between the plurality of reservoir datums as inputs and the at least one fracking simulation model output as outputs, thereby eliminating a need for executing the fracking simulation model during real-time operations.

15 . The method of claim 10 , wherein the fracking production parameter comprises at least one of: optimal drilling depth, optimal well location, optimal proximity of fractures, optimal quantity of fractures, optimal fracking fluid pressure, and optimal fracking fluid flow rates.

16 . The method of claim 10 , wherein the fracking simulation model is configured to calculate fracture propagation using at least one fracture mechanics criterion, wherein the at least one fracture mechanics criterion comprises an evaluation of a plurality of reservoir datums.

17 . The method of claim 10 , wherein the fracking simulation model is configured to simulate fluid flow in a fracture by implementing a fluid dynamics model that relates fluid parameters and downhole conditions to pressure gradient along the fracture, and wherein the fracking optimization machine learning model is trained to predict this fluid flow without executing the fracking simulation model.

18 . The method of claim 10 , wherein the fracking simulation model is configured to calculate rock deformation as a function of reservoir datum using a rock behavior model, and wherein the fracking optimization machine learning model is trained to predict this rock deformation without executing the fracking simulation model.

19 . A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations for fracking optimization, the operations comprising:

receiving a plurality of reservoir datums from a sensor suite, wherein the plurality of reservoir datums comprises:

at least one surface reservoir datum detected using a first sensor of the sensor suite; and

at least one downhole reservoir datum detected using a second sensor of the sensor suite;

generating production training data as a function of the plurality of reservoir datums, wherein generating the production training data comprises:

processing the at least one downhole reservoir datum into a standardized data format using a data conversion module;

obtaining at least one fracking simulation model output, wherein the fracking simulation model is configured to calculate at least one of: rock deformation, reservoir pressure, fracture propagation, fluid flow in a fracture; and

correlating the plurality of reservoir datums with the at least one fracking simulation model output to generate the production training data;

training a fracking optimization machine learning model using the production training data, wherein the fracking optimization machine learning model is configured to predict the at least one fracking simulation model output without requiring execution of the fracking simulation model;

receiving a surface target reservoir datum from a sensor suite; and

determining a fracking production parameter in real-time using the surface target reservoir datum and the trained fracking optimization machine learning model without requiring execution of the fracking simulation model.

20 . The non-transitory computer-readable storage medium of claim 19 , wherein the operations further comprise:

predicting, using the fracking optimization machine learning model, a likelihood of fracturing into a nearby wellbore based on the surface target reservoir datum;

generating a well spacing plan based on the fracking production parameter determined using the fracking optimization machine learning model;

wherein the fracking optimization machine learning model is configured to predict at least one of fracture height, fracture width, fracture length, and fracture conductivity as the at least one fracking simulation model output;

wherein the fracking production parameter comprises at least one of optimal drilling depth, optimal well location, optimal proximity of fractures, optimal quantity of fractures, optimal fracking fluid pressure, and optimal fracking fluid flow rates;

wherein the fracking simulation model is configured to calculate fracture propagation using at least one fracture mechanics criterion, simulate fluid flow in a fracture by implementing a fluid dynamics model that relates fluid parameters and downhole conditions to pressure gradient along the fracture, and calculate rock deformation as a function of reservoir datum using a rock behavior model; and

wherein the fracking optimization machine learning model is trained to predict the fracture propagation, fluid flow, and rock deformation without executing the fracking simulation model, thereby eliminating a need for executing the fracking simulation model during real-time operations.