DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Deep reinforcement learning control of hydraulic fracturing

Journal Article · · Computers and Chemical Engineering
 [1];  [2]
  1. Texas A & M University, College Station, TX (United States); OSTI
  2. Texas A & M University, College Station, TX (United States)

Hydraulic fracturing is a technique to extract oil and gas from shale formations, and obtaining a uniform proppant concentration along the fracture is key to its productivity. Recently, various model predictive control schemes have been proposed to achieve this objective. But such controllers require an accurate and computationally efficient model which is difficult to obtain given the complexity of the process and uncertainties in the rock formation properties. In this article, we design a model-free data-based reinforcement learning controller which learns an optimal control policy through interactions with the process. Deep reinforcement learning (DRL) controller is based on the Deep Deterministic Policy Gradient algorithm that combines Deep-Q-network with actor-critic framework. In addition, we utilize dimensionality reduction and transfer learning to quicken the learning process. We show that the controller learns an optimal policy to obtain uniform proppant concentration despite the complex nature of the process while satisfying various input constraints.

Research Organization:
Texas A & M University, College Station, TX (United States)
Sponsoring Organization:
National Science Foundation (NSF); Texas A&M Energy Institute; USDOE Office of Energy Efficiency and Renewable Energy (EERE)
Grant/Contract Number:
EE0007888
OSTI ID:
1977009
Journal Information:
Computers and Chemical Engineering, Journal Name: Computers and Chemical Engineering Journal Issue: C Vol. 154; ISSN 0098-1354
Publisher:
ElsevierCopyright Statement
Country of Publication:
United States
Language:
English

References (30)

Temporal clustering for order reduction of nonlinear parabolic PDE systems with time‐dependent spatial domains: Application to a hydraulic fracturing process journal April 2017
Toward self‐driving processes: A deep reinforcement learning approach to control journal June 2019
Online actor–critic algorithm to solve the continuous-time infinite horizon optimal control problem journal May 2010
Approximate dynamic programming based approach to process control and scheduling journal September 2006
Optimal pumping schedule design to achieve a uniform proppant concentration level in hydraulic fracturing journal June 2017
Development of local dynamic mode decomposition with control: Application to model predictive control of hydraulic fracturing journal November 2017
Modeling of hydraulic fracturing and designing of online pumping schedules to achieve uniform proppant concentration in conventional oil reservoirs journal June 2018
Model order reduction of nonlinear parabolic PDE systems with moving boundaries using sparse proper orthogonal decomposition: Application to hydraulic fracturing journal April 2018
Data-driven identification of interpretable reduced-order models using sparse regression journal November 2018
Reinforcement Learning – Overview of recent progress and implications for process control journal August 2019
Deep hybrid modeling of chemical process: Application to hydraulic fracturing journal March 2020
Reinforcement learning based optimal control of batch processes using Monte-Carlo deep deterministic policy gradient with phase segmentation journal January 2021
Process systems engineering – The generation next? journal April 2021
Computer simulation of hydraulic fractures journal July 2007
Continuous control of a polymerization system with deep reinforcement learning journal March 2019
A model-based deep reinforcement learning method applied to finite-horizon optimal control of nonlinear control-affine system journal March 2020
Neural network approach to continuous-time direct adaptive optimal control for partially unknown nonlinear systems journal April 2009
Handling Spatial Heterogeneity in Reservoir Parameters Using Proper Orthogonal Decomposition Based Ensemble Kalman Filter for Model-Based Feedback Control of Hydraulic Fracturing journal February 2018
Enlarging the Domain of Attraction of the Local Dynamic Mode Decomposition with Control Technique: Application to Hydraulic Fracturing journal March 2019
Evaluating the Performance of a Fracturing Treatment Design journal June 2014
Model-Based Closed-Loop Control of the Hydraulic Fracturing Process journal January 2015
Human-level control through deep reinforcement learning journal February 2015
Mastering the game of Go with deep neural networks and tree search journal January 2016
Mastering the game of Go without human knowledge journal October 2017
Temporal difference learning and TD-Gammon journal March 1995
Determination of Proppant and Fluid Schedules From Fracturing-Pressure Decline journal July 1986
Experimental and Numerical Modeling of Convective Proppant Transport(includes associated papers 31036 and 31068 ) journal March 1995
Propagation of a Vertical Hydraulic Fracture journal August 1972
Numerical Solution of Sand Transport in Hydraulic Fracturing journal January 1978
Widths of Hydraulic Fractures journal September 1961