Artificial intelligence (AI) has been used for more than two decades as a development tool for solutions in several areas of the E&P industry: virtual sensing, production control and optimization, forecasting, and simulation, among many others. Nevertheless, AI applications have not been consolidated as standard solutions in the industry, and most common applications of AI still are case studies and pilot projects.
An analysis of a survey conducted on a broad group of professionals related to several E&P operations and service companies will be presented. The survey captures the level of AI knowledge in the industry, the most common application areas, and the expectations of the users from AI-based solutions. It also includes a literature review of technical papers related to AI applications and trends in the market and R&D.
The survey helped to verify that (a) data mining and neural networks are by far the most popular AI technologies used in the industry; (b) approximately 50% of respondents declared they were somehow engaged in applying workflow automation, automatic process control, rule-based case reasoning, data mining, proxy models, and virtual environments; (c) production is the area most impacted by the applications of AI technologies; (d) the perceived level of available literature and public knowledge of AI technologies is generally low; and (e) although availability of information is generally low, it is not perceived equally among different roles.
To illustrate the applications of AI in the E&P industry, a case study is presented to predict liquid rate and water cut performance using neural networks (NN) in a mature reservoir with more than 20% water cut. Real-time surveillance and monitoring of production operation processes has proven to be operationally and economically important for managing complex, high-cost reservoirs. However, predicting short-term production levels and production upsets, for instance related to pump settings, has posed a tremendous challenge for the oil and gas industry. While operators routinely forecast production for the next month, sophisticated tools such as full field numerical simulation models are of limited use in predicting very short-term production. Similarly, while nodal analysis can estimate current operating conditions, it cannot be used for prediction. Due to its simplicity, rapid training and demonstrated results, the NN technique has emerged as a potential solution that can predict short-term well production behavior with acceptable accuracy.
The NN was trained using available surface and down-hole real-time production data, time-dependent data, and completion designs. The time-dependent data are included as time series’ configured to let users generate scenarios by changing well operations. This approach not only provided a base case prediction, but also a fast simulation tool to predict scenarios after making what-if’s in control variables such as Tubing Head Pressure (THP) and Pump Frequency, which let the user to predict and circumvent negative well pump events. The NN results may be used as a Virtual Production Meter that can show better results than traditional techniques such as well models based on pure physics.
Dr. Luigi Saputelli
Dr. Saputelli is a petroleum engineer with 20+ years’ experience in reservoir engineering, field development, production engineering, drilling engineering, production operations and oilfield automation projects. He holds a PhD in Chemical Engineering (2003) from the University of Houston, a M.Sc. in Petroleum Engineering (1996) from Imperial College, London, and a B.Sc. in Electronic Engineering (1990) from Universidad Simon Bolivar, Caracas, Venezuela
He worked as Production Engineering Senior Advisor at Hess Corporation (2009-2012), Halliburton (2003-2009) where he acted as technical lead of various major projects such as Petrobras’ Barracuda-Caratinga fields Real Time Operations project (2005-2006), PDVSA integrated modeling and field exploitation plans for Carito and Orocual fields (2006-2007). Dr. Saputelli later acted as Production Operations Regional Practice Manager and Field Development Global Practice Manager for Halliburton; also worked in PDVSA E&P (1990-2001) and acted as a well planning senior advisor, production technologist, reservoir modeling and simulation engineer.
He has worked in various countries: Venezuela, Argentina, Brazil, Colombia, Saudi Arabia, Kuwait, Arab Emirates, Nigeria, Thailand, Malaysia, Indonesia, England, Scotland and USA. He is an industry recognized researcher, lecturer, SPE Liaison and member of various committees. He has published over 50 industry papers and has three patents on applied technologies for reservoir management, real time optimization and production operations. He is currently President of Frontender Corporation and engaged in various Digital Oilfield implementations.