The use of artificial lift equipment for oil production in onshore reservoirs is becoming increasingly more important to help sustain the production rates of declining oil fields. Oil field producers therefore depend on the efficient operation of the artificial lift equipped and it is becoming increasingly more important to ensure maximum uptime of the equipment for the continuous production of oil.
One of the widely used methods for artificial lift is using Electrical Submersible Pumps to produce high volumes from deep oil wells. The ESP is a very effective method of artificial lift due to its unique characteristic of having the complete pump assembly and electrical motor submersed directly in the well fluid. This however requires a complex technical design of the pump and electrical motor to ensure safe operation several thousand feed below surface. It is therefore necessary to implement systems that can monitor the pump operation and notify the operator of events that will result in failure of the equipment.
Typically, ESPs are connected to SCADA or other distributed control systems to provide supervisory and control functions for their effective operation as well as for operational visibility. Today, many diagnostic methods are available to determine the health and status of an ESP system by making use of that functionality in its automation system. However, while these methods can provide insightful analysis of problems, they usually require constant monitoring of a human operator who is able to react in time to alarm notifications or implement corrective action. The correct operation of the ESP largely depends on the decisions made by the ESP field operator and his ability to effectively control the ESP fleet based on his experience. The complexity of the operator’s task increases with the size of the of ESP fleet that the operator must manage at any given point in time.
But this situation is changing, with efforts being made to reduce the dependency on the human operator by implementing digital support systems. With recent advances in artificial intelligence (AI) combined with the new Internet of Things (IoT) technologies, it is possible to effectively use data-driven analytics fueled by large data sets to assist the operator with the task of operating ESP fleets. In particular, AI technology that involves deep learning and neural networks can be extremely effective in detecting abnormal behavior of complex physical systems such as ESPs, based on the data gathered from the system, providing decision support for remediating or managing the causative issues.
One of the primary advantages of using AI technology is its ability to detect abnormal behavior in complex systems. Such an AI system can be implemented to monitor ESP systems using the real-time process data from the Supervisory system and then using a neural network model identify abnormal ESP pump behavior. The paper discuss how such an AI based anomaly detection systems can be used in a extended form to implemented an autonomous surveillance system which can monitor and entire ESP fleet. The purpose of the autonomous surveillance system is to support the operator in his supervisory tasks by doing the selection and prioritization of ESP units that requires operator attention.
This paper is a continuation of an earlier paper which discussed the possibility to implement a predictive maintenance system for ESPs using AI. This paper further elaborates the implementation of an autonomous surveillance solution for ESP systems using the predictive maintenance solution and explain how it can be implemented using AI technology in combination with a cloud-based IoT platform.