Artificial Intelligence Optimizes Oil and Gas Production
An AI-based application enabled operators to preempt ESP failures while optimizing production.
The Internet of Things (IoT), along with advances in sensor technology, data analytics, and artificial intelligence (AI), has paved the way for significant efficiency and productivity gains in the oil and gas industry. One application in particular, electrical submersible pumps (ESPs), has benefited from these advances. The complete paper explores an AI-based application that enables operators to preempt costly ESP failures while optimizing production.
AI and Machine Learning
Cloud-based platforms can provide offshore operators with access to advanced analytics software featuring AI algorithms that analyze incoming data for anomalies that could signal trouble ahead in the monitored equipment. This type of AI capability is often referred to as machine learning (ML). ML uses statistical analysis to infer the probability of a hypothesized event, combining this capability with neural network models to learn the relational behavior of a system and its components. AI-based ML algorithms, once authored and deployed, first need to be trained with sample baseline operating data and then provided with actual operating data.
Using AI To Forecast ESP Operational Issues
ESP diagnostic systems typically use two methods, physics-based modeling and expert modeling, to ascertain operating issues. In the former, mathematical equations describe the physical behavior of a process. Although this model can predict ESP behaviors accurately, the development of the equations requires expensive, highly skilled effort. Also, physics-based modeling can be limited in its applicability across a fleet of similar systems operating with slightly varied parameters owing to different environmental and geological conditions.
Expert modeling costs less to develop because its analytical model does not require the same level of sophisticated skills, and its results are more easily interpreted. However, expert modeling has the disadvantage of having expert biases built in so that it tends to discover known anomalies in equipment behavior, overlooking unknown anomalies that can still disrupt an ESP’s operation or cause it to fail.
ML helps to overcome this problem. ML starts with a programmed neural network model, then uses historical equipment data to teach the model how a system and its constituent parts should behave in terms of the system’s as-designed and as-tested operating parameters. In effect, the more data that are processed, the smarter the model becomes.
A challenge with an ML-based monitoring and diagnostics model is that it requires a data training set, which needs significant time and effort to prepare. Nonetheless, once the ML model has learned the normal behavioral relations across all of the varied parameters of a system, such as an ESP, it can begin comparing massive amounts of system data to its baseline data set in near real time. Using pattern recognition, it can uncover any anomalous relationships that emerge in a system’s operation, and then analyze those differences and provide the probabilities of future behavior.
Several caveats are in order with this description of ML-based monitoring and diagnostics models. First, operators must be able to effectively manage vast amounts of process data. Second, to ensure data integrity, proper data structuring is critical. Third, data availability and its time synchronization are important to ensure the fidelity of the data as input to the ML-based model.
Notably, especially in offshore applications, ML-based models typically are based in cloud platforms where the advanced analytics software resides and does its work. Data in motion are encrypted and transmitted by IoT connectivity, either through wired or wireless media. Public cloud platforms can be as secure, or more secure, than on-premise facilities, and offer pay-as-one-goes subscriptions that conserve capital and time. Building an on-premise infrastructure can take months, while creating a cloud-based solution can be achieved in hours or days.
ML Deployed for a Predictive ESP Maintenance Model
In a proof-of-concept case and a later pilot, ML was used to develop a probabilistic, predictive maintenance model that was implemented on an IoT-based cloud platform for a fleet of 30 ESPs operating in an onshore oil field at various depths. The ESP pumps varied from 200–500 kW in power, driven from an aboveground variable-frequency drive (VFD) with a medium-voltage input drawn from a local electric utility. The architecture of the system is shown in Fig. 1. The components included the following:
ESP. Each ESP operated with belowground instrumentation to monitor pump parameters. VFD instrumentation also gathered data from the pumps along with other wellsite instrumentation used to monitor additional operational parameters.
Supervisory Control and Data-Acquisition (SCADA) System. A centralized SCADA system linked the ESPs, providing supervisory control of the fleet’s pumps. The system fed the process data to a high-speed historian database.
Fiber-Optic Network. The SCADA system was connected permanently to the distributed ESP systems by a high-availability, high-speed, fiber-optic network.
Monitoring Application. The ML-based monitoring application was developed with an open-source software library.
Predictive Maintenance Application. The Web-based predictive maintenance application was coded in hypertext markup language (HTML) and deployed in the same public cloud as the monitoring application.
Preparing the Baseline Training-Data Set
To develop a common operating model of the complex ESP systems used in this project, sufficient high-quality data were required over a long-enough time period to reasonably represent an ESP’s typical operation. After extracting these data sets, the project’s data scientists conducted manual data cleansing and evaluation, resulting in an optimized baseline training data set for the ML predictive maintenance model. The last step in this preparatory phase of the project was to develop a generic, multilayer, neural network with the ML monitoring application, and then apply the different training sets to train each of the neural networks corresponding to their ESP variant counterpart.
Using Historical Data for a Proof of Concept
Although the project hypothesized that an ML-based, probabilistic predictive maintenance model for a sizeable ESP fleet was possible, the first step was to prove the concept. For this important validation, data pulled from the SCADA historian database for identical periods of an ESP’s operations were used. This included normal operating data as well as data associated with ESP failures. That became the data-source input to the ML-based model, which was expected to conduct analytics on the data, so that time periods of normal ESP operation could be distinguished from time periods with anomalous ESP behaviors.
Once the ML-based model began processing the unfiltered historical data of the ESP fleet’s variant groups, it showed that it could differentiate normal and abnormal behaviors over comparable time periods. Numerous experiments were performed with the ML model by using the same historical data as input and then adjusting different tuning parameters of the monitoring application that used the ML technology. One of the key parameters tested this way was an anomaly threshold parameter. Shown as a percentage value, this is a cumulative indication, across all process variables, of the probability that the ML-based model will predict an anomaly. The test of this parameter was to determine its relative effect on the timing and accuracy of the model’s advance notification for any specific anomaly. Indeed, the anomaly-threshold parameter did affect the accuracy and timing of the predictions. When set to a high value such as greater than 80%, the ML-based model would predict anomalies before an ESP’s failure inside short notification periods accurately, usually in the range of hours. Conversely, if the anomaly threshold was set to a lower value, such as less than 40%, the ML-based model still would detect data anomalies accurately but within a much earlier notification period, in the range of several days.
Fortunately, the model could be tuned finely to balance between providing early predictions with longer notification periods without generating large numbers of false anomaly predictions. In one test, the ML-based model discovered significant anomalies 12 days ahead of one ESP’s failure. The predicted anomalies were analyzed and verified with data from the failed ESP’s historical records.
Pilot Phase Using Real-Time Production Data
After the successful proof of concept, the ESP fleet operator agreed to conduct a pilot-project version of the ML-based ESP predictive maintenance solution with anomaly detection. The goal was to test its deployment and performance in a production environment. The pilot was performed with the same system architecture illustrated in Fig. 1.
In the pilot, real-time production data was transmitted directly from the ESP fleet’s central SCADA system. From there, it fed into the ML-based predictive maintenance model through highly secure IoT connectivity to a public cloud platform where the analytics applications resided. The latter processed the ESP operating data as input, analyzing it in real time to identify anomalous behaviors that could be mitigated or remediated by appropriate preventative maintenance. The pilot project again verified the use of ML as the basis for an ESP predictive maintenance model. At that point, the ESP fleet operator deployed it as a supplementary expert system with no additional integration with the central SCADA system.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper OTC 29384, Applying Artificial Intelligence To Optimize Oil and Gas Production, by Christoph Kandziora, Siemens, prepared for the 2019 Offshore Technology Conference, Houston, 6–9 May. The paper has not been peer reviewed.