_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 230613, “Transforming Hydraulic Fracturing: The First-Ever Closed-Loop Completions Program, ” by Awais Navaiz, SPE, and Price F. Stark, SPE, Halliburton, and Matt Paradeis, SPE, Chevron, et al. The paper has not been peer-reviewed. _ The primary goal of this work is to present the design and successful field deployment of the first closed-loop hydraulic fracturing program. This system performs fully autonomous fracturing operations, adjusting completion parameters in real time based on subsurface feedback while also condensing the decision-making lag from minutes or hours to a few seconds. The core of this closed-loop system consists of three components: sensing, decision logic, and execution layer. A conditional workflow was invented to tie these individual layers to observe subsurface diagnostic data, decide the appropriate actions, and trigger the fleet to execute these adjustments, all without any manual intervention. Introduction Surface-level automation for fracturing operations has gained traction rapidly with the introduction of electric-powered fleets, allowing precise control of pumping equipment, leading to repeatable execution, increased efficiency, and decreased risk of screenouts. The capability to have autonomously controlled fleets serves as the execution bedrock of this program. Building on this foundation, the present work expands this autonomy from execution to optimization by enabling dynamic formation responses to drive surface equipment without human intervention. Planning and Deployment For this discussion, a fully closed-loop operation is defined as a self-regulating, self-contained process that uses feedback to control its performance. The process is considered fully closed-loop when the feedback is integrated continuously into the system without requiring manual intervention. Decisions would have to be received and implemented reliably at the edge. Field operations are fast-paced with limited communication and require dozens of individuals to work in cohesion around the clock. Introducing another layer of highly dynamic, operational procedures into an already-complex environment results in adjustments being missed and lacks consistency. It was clear that the only way to successfully deploy this decision-making dynamically was through automation of the entire fracturing fleet. Subsurface feedback had to be interpreted live and translated into actions that drove desired outcomes. This uncharted territory warranted deep-rooted collaboration between the operator and the service company. A shared design framework was collaboratively established that allowed for building tailored workflows. To benchmark performance and provide statistical insight for decision-making, a large and consistent data set of subsurface diagnostic information had to be aggregated instead of isolated samples. The deployment was subdivided into the following three major phases: - Baseline acquisition: Deploy nonintrusive diagnostic tools to monitor subsurface data continuously in the ongoing completions program - Efficiency optimization: Automate the control of fracturing treatments to enable consistency in execution while lowering cycle times and facilitating a low-latency infrastructure for rapid, logic-driven adjustments - Closed loop: Integrate subsurface feedback continuously with surface automation to execute treatment changes autonomously
Chris Carpenter (Mon,) studied this question.