Developing an ESP Lifespan Predictive Model Using Artificial Intelligence: A Case Study On an Omani Oilfield

Abstract

The Electrical Submersible Pump (ESP) is the most effective and consistent artificial lift method for medium to high production rates. Although the capital cost of ESP is high, it pales in comparison to the production losses resulting from its failure. Recently, Machine Learning (ML) has gained significant attention in the oil and gas industry due to its predictive power. This paper aims to develop a ML model to predict ESP lifespan and identify the key features that influence its longevity. The study reviewed the failure history of more than 100 wells from an Omani oilfield, with 132 ESP failures attributed to sand and scale accumulation. The dataset includes 36 static features related to ESP design, installation, commissioning, failure, pull-out, and teardown. Three algorithms, namely Support Vector Regressor (SVR), Random Forest Regressor (RFR), and Extreme Gradient Boosting Regressor (XGBR), were selected. Hundreds of tests were performed on each algorithm to optimize the parameters and hyperparameters, based on mean absolute error, average residual, and determination coefficient. The study developed a model with two levels to predict the lifespan of ESP before installation and after the last valid well test. The model had a mean absolute error of 25 days and 8 days for the first and second levels, respectively, with a determination coefficient of 60% and 73%. The model showed that certain factors related to pump and motor design have the most significant impact on the longevity of the ESP before installation. Pump discharge pressure and flow rates of oil and water are crucial to monitor and control during its operational lifespan. The findings emphasize the importance of careful selection and design of ESP components to ensure a long-lasting lifespan. By scheduling ESP maintenance before failure, these findings can help mitigate capital costs, while preparing the necessary hoist, rig, and materials for ESP replacement can avoid deferred operational costs.