In this paper, we present an automated data-driven workflow using Machine Learning (ML) for gas lift optimization in unconventional fields. This workflow integrates a ML model that accurately forecasts the Gas Lift Performance Curve, and a Bayesian Optimization Framework to solve for the optimal gas injection rates under the constraints of facility capacity. The ML model leverages the historical production time series data without requiring downhole gauges or multi-rate well tests. We piloted this workflow on 30 wells across 5 well pads in Bakken and obtained >5% production uplift on average. With the success of the pilot, we have now fully-deployed this workflow in Bakken across 200+ gas lift and plunger-assisted gas lift (PAGL) wells. Moreover, the ML-based gas lift optimization workflow presented in this paper is an effective and economic solution for other assets where downhole data or multi-rate testing are not available/feasible due to cost or facility constraints.
For unconventional fields, oil rates typically drop sharply after a few months of production in naturally flowing wells due to the rapid decline of the reservoir pressure. Different kinds of artificial lift methods are then required to continue the production. Gas Lift (GL) is one of the most economic and commonly used artificial lift methods in unconventional fields. During a gas-lift operation, gas is injected into the tubing through a valve from the annulus. The gas mixing with the fluid reduces the mixture density, which decreases the hydrostatic pressure gradient, lowers the flowing bottom hole pressure, increases the pressure draw down and thus ultimately increases the oil flow rate (Brown 1967). When a GL well is in a semi-steady state, its production rate versus gas injection rate typically follows a bell-shaped curve called "Gas Lift Performance Curve" (Figure 1). At a relatively high gas injection rate, the frictional pressure loss starts to compete with the hydrostatic head reduction, resulting in lower production rate. Moreover, the Gas Lift Performance Curve evolves over time as well matures. Therefore, one of the key business questions is to find the optimal gas injection rate that yields the maximum liquid production rate for any given GL well at any time in its life span.