Hydraulic fracturing job plan real-time revisions utilizing detected response feature data

Patent metadata

Field	Value
Publication number	US-2022027538-A1
Application number	US-201817297335-A
Country	US
Kind code	A1
Filing date	Dec 27, 2018
Priority date	Dec 27, 2018
Publication date	Jan 27, 2022
Grant date	—

Abstract

The disclosure is directed to methods to design and revise hydraulic fracturing (HF) job plans. The methods can utilize one or more data sources from public, proprietary, confidential, and historical sources. The methods can build mathematical, statistical, machine learning, neural network, and deep learning models to predict production outcomes based on the data source inputs. In some aspects, the data sources are processed, quality checked, and combined into composite data sources. In some aspects, ensemble modeling techniques can be applied to combine multiple data sources and multiple models. In some aspects, response features can be utilized as data inputs into the modeling process. In some aspects, time-series extracted features can be utilized as data inputs into the modeling process. In some aspects, the methods can be used to build a HF job plan prior to the start of work at a well site. In other aspects, the methods can be used to revise an existing HF job plan in real-time, such as after a treatment cycle, a pumping stage, or a time interval.

First claim

1 . A method to design a hydraulic fracturing (HF) job plan to direct operations of well site equipment for a well, comprising:
extracting a response feature set from time-series pumping data gathered during pumping operations or during shutdown operations; building a prediction model for potential well production, utilizing a mathematical model or statistical model, and wherein the prediction model utilizes the response feature set; identifying a combination of features from the response feature set to satisfy one or more key performance indicators (KPI); receiving a well site data set, where the well site data set is from one or more treatment cycles of the well; and revising the HF job plan utilizing the well site data set and the prediction model, wherein the combination of features fail to satisfy the KPI.

2 . The method as recited in claim 1 , wherein the response feature set is one or more of a pressure response data, flow distribution data, microseismic response data, and fracture dimension data.

3 . The method as recited in claim 1 , wherein the response feature set is one or more of first and second derivative of pressure with time, non-linear spline fits of pressure responses, first, second, and third moment during pad phase, proppant pumping, diversion cycle properties, and shape of entire treatment cycles.

4 . The method as recited in claim 1 , wherein the response feature set for the shutdown operations are one or more of instantaneous shut-in pressure, pressure decline slope post pumping, G-function parameters, and water hammer amplitude, frequency, and decay.

5 . The method as recited in claim 1 , wherein the KPI is one or more of well productivity, total cost of ownership, return on investment, and fracture dimension.

6 . The method as recited in claim 1 , wherein the response feature set includes a response feature from historical KPI and time-series data from one or more past executed jobs.

7 . The method as recited in claim 1 , wherein the extracting further comprises:
aggregating uncontrollable features, wherein the uncontrollable features are uncontrollable for HF jobs; aggregating controllable features, wherein the controllable features are controllable for HF jobs; and the response feature set comprises the uncontrollable features and the controllable features.

8 . The method as recited in claim 7 , wherein the identifying further comprises:
determining the combination of features utilizing the prediction model, wherein the prediction model utilizes the response feature set and the controllable feature set to determine the KPI impact.

9 . The method as recited in claim 7 , wherein the revising further comprises:
determining a subset of controllable features to be adjusted in the HF job plan to improve satisfaction of the KPI.

10 . The method as recited in claim 7 , wherein the aggregating uncontrollable features and the aggregating controllable features are aggregated over one of the treatment cycle, a diversion cycle, a pumping stage, or a time-interval.

11 . The method as recited in claim 1 , wherein the receiving is real-time HF job data.

12 . The method as recited in claim 1 , wherein the extracting and the building utilize an ensemble model utilizing a single stage predictive model or a multiple stage predictive model, and wherein the ensemble model consolidates one or more modeling techniques and data sources.

13 . (canceled)

14 . (canceled)

15 . The method as recited in claim 1 , wherein the extracting the response feature set further comprises:
receiving a first time-series pumping data set, wherein the first time-series pumping data set comprises one or more of treating pressure, slurry rate, and proppant concentration; receiving a second time-series pumping data set, wherein the second time-series pumping data set comprises one or more of treating pressure, slurry rate, and proppant concentration; identifying a first event set, wherein the first event set comprises event time intervals, utilizing the first time-series pumping data set; training a machine learning model utilizing the first event set; and estimating a second event set, utilizing the second time-series pumping data set and the machine learning model.

16 . The method as recited in claim 15 , wherein the first event set and the second event set further comprise event property data.

17 . The method as recited in claim 15 , wherein the first event set and the second event set comprise event types of one or more of treatments, diversion cycles, san slug, minifrac, step-up, step-down, instantaneous shut-in pressure, breakdown, and screenouts.

18 . The method as recited in claim 15 , wherein the first time-series pumping data set and the second time-series pumping data further comprise user defined data, and wherein the first event set and the second event set further comprise user defined data.

19 . A computer program product having a series of operating instructions stored on a non-transitory computer-readable medium that directs a data processing apparatus when executed thereby to perform operations to design a hydraulic fracturing (HF) job plan to direct operations of well site equipment of a well, having operations comprising:
extracting a response feature set from time-series pumping data, wherein the time-series pumping data is gathered during pumping operations or during shutdown operations; building a prediction model for potential well production, utilizing a mathematical model or statistical model, and wherein the prediction model utilizes the response feature set; identifying a combination of features from the response feature set to satisfy one or more key performance indicators (KPI); receiving a well site data set, where the well site data set is from one or more treatment cycles of the well; and revising the HF job plan utilizing the well site data set and the prediction model, wherein the combination of features fail to satisfy the KPI.

20 . The computer program product as recited in claim 19 , wherein the extracting further comprises:
aggregating uncontrollable features, wherein the uncontrollable features are uncontrollable for HF jobs; aggregating controllable features, wherein the controllable features are controllable for HF jobs; and the response feature set comprises the uncontrollable features and the controllable features.

21 . The computer program product as recited in claim 20 , wherein the identifying further comprises:
determining the combination of features utilizing the prediction model, wherein the prediction model utilizes the response feature set and the controllable feature set to determine the KPI impact.

22 . The computer program product as recited in claim 20 , wherein the revising further comprises:
determining a subset of controllable features to be adjusted in the HF job plan to improve satisfaction of the KPI.

23 . The computer program product as recited in claim 19 , wherein the extracting the response feature set further comprises:
receiving a first time-series pumping data set, wherein the first time-series pumping data set comprises one or more of treating pressure, slurry rate, and proppant concentration; receiving a second time-series pumping data set, wherein the second time-series pumping data set comprises one or more of treating pressure, slurry rate, and proppant concentration; identifying a first event set, wherein the first event set comprises eve

Assignees

• Halliburton Energy Services Inc

Inventors

• Walters Harold Grayson
• Fulton Dwight David
• Bhardwaj Manisha
• Potty Ajish Sreeni Radhakrishnan

Classifications

• G06N20/20 Primary

  Ensemble learning · CPC title

• E21B43/26

  by forming crevices or fractures · CPC title

• E21B2200/20

  Computer models or simulations, e.g. for reservoirs under production, drill bits · CPC title

• G06F30/28 Primary

  using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD] · CPC title