A Data-Based Reliability Analysis of ESP Failures in Oil Production Wells

Electrical Submersible Pumps (ESPs) are one of the most widely used artificial lift methods in the petroleum industry. However, ESP failures are unanticipated and common occurrences with significant financial impacts for the operators. Analysis of the ESP performance and failures are essential in its design and optimization. This paper presents a statistical approach for diagnosing and evaluating the root causes of ESP failures. The analysis is based on the field data gathered from the surface and downhole ESP monitoring equipment over five years of production of 10 wells. Electrical failures are the most common general cause of ESP failures, accounting for 61% of all failures, followed by motor failure and gas locking. Specifically, power failure, under-voltage, voltage unbalance, and motor underload are the most common occurrences. The data trends are analyzed for the two weeks before each specific failure, and conclusions are made on the warning signs to predict failures. In addition, a Weibull statistical analysis model is constructed to evaluate the reliability features and estimate the ESP failure probability, allowing operators to perform preventive maintenance. The results provide guidelines for ESP operations and contribute to reducing or preventing ESP downtimes and operating costs.