An oil and gas well produces mixed streams of intermingled hydrocarbons and water that may include liquids and gases. The market value of these hydrocarbon streams varies widely; lighter streams such as natural gas and natural gas liquids are often less valuable than heavier crude oil streams that may be refined into jet fuel or naphtha products, such as gasoline. Oil and gas well operators have significant economic interest in understanding the volume of each type of hydrocarbon phase in a gathering pipeline as well as the amount of water, but this is not easy to determine. Producing wells may change their outputs frequently, and gathering systems attached to multiple wells may see shifts in total and relative volume of intermingled phases of hydrocarbons and water on a regular basis.
For many years, oil and gas operators have obtained measurements of hydrocarbon volumes through the use of multiphase flowmeters (MPFMs). These sophisticated devices typically estimate the oil, gas and water composition by combining total flow measurements with density measurements from a gamma densitometer. This technology has advanced over the years, but it remains expensive, static and, because of its bulk and radioactive components, it may require significant infrastructure investments and often unavailable real estate to collect and analyze data.
The growing technology forces of increased sensor availability, cheaper data storage and computing capacity have resulted in more opportunities to apply advanced analytical methods in new ways across industries and business processes. In particular, these factors combine to create new opportunities for advanced analytical software to augment or even replace hardware functions in several industrial operations. This is true for MPFMs as well.
ABB and Arundo Analytics collaborated to create a virtual multiphase flowmeter (VMPFM) based on ABB’s experience in hydrocarbon flow control and monitoring, and Arundo’s expertise in transforming complex analytical models into enterprise-scale, web-accessible software. This VMPFM is designed to enable the core business proposition of a traditional MPFM — measuring component volumes of intermingled hydrocarbon phases — at a lower cost in initial capital investment and ongoing operating costs. The VMPFM analyzes multiphase flow in any well or pipeline based on sensor measurements. The analysis combines information about well or pipeline configurations with advanced physical models and numerical solvers in a cloud-based software environment. End users may access results through any web-enabled device or into local control systems via a web-based application program interface.
Of course, such a software-based system still requires data from physical measurements to produce accurate results. A piece of software sitting on a local gateway device — a rugged, field-deployable computer with internet capabilities — may be used to sample, spool and intelligently stream data from the field to the cloud. This software can connect to industrial control systems or data historians, or directly to instrumentation via common industrial protocols such as OPC-UA, OPC-DA, Profinet or Modbus. Depending on internet connectivity and business requirements for data refresh, data can be buffered for bandwidth optimization, analyzed locally or, depending on the local computing capabilities, used in full-fledged advanced analytical models. Most often, data are sent from the field to the cloud at intervals ranging from one to five minutes.
The functional core of the new offering is actually not new at all; the analytical models and optimizer that form it have been used by one of the largest oil companies in the world for more than 10 years. In these installations in which the technology is implemented on servers at the actual offshore facilities, the technology is used for monitoring and optimizing flow from the wells and for managing the safe injection of water into the reservoir to optimize pressure support and thereby production. The combined offering is designed to provide accuracy with significantly faster, more flexible and cost-efficient deployment.
The advantages of using virtual flowmeters are numerous. By substituting software for expensive physical MPFMs, operators avoid significant capital and operational expenses. In addition, the quality of the analysis may improve since data can be streamed back on an ongoing basis, creating a more coherent picture of pipeline activity. Furthermore, operators can analyze data related to measurements from sensors at any point in the system rather than from a few physical sampling points. All of this creates a better understanding of flow volumes and improves the business decision-making processes informed by these analytics.
The use of advanced artificial intelligence algorithms will further improve the information presented by VMPFMs. By combining information across all wells of a production field, production can be optimized in the field rather than in individual wells. Machine learning can also be used to automate the configuration of the VMPFMs, thereby easing implementation and scalability.
Mogens Mathiesen is co-founder of Arundo Analytics, which enables advanced analytics in heavy industries. He holds a Master of Science in ocean engineering and marine cybernetics from the Norwegian University of Science and Technology and was a Ph.D. candidate at the University of California, Berkeley, and Imperial College, London, where he focused on offshore oil and gas production wireless control. Mathiesen can be reached at firstname.lastname@example.org.
Amitav Misra is VP marketing at Arundo. He works with industrial equipment manufacturers and major asset owners on advanced analytics solutions. Misra holds a Bachelor of Arts in economics from Stanford University. He can be reached at email@example.com.
Espen Storkaas leads ABB´s global digitalization business for upstream oil and gas. He holds a Master of Science and Ph.D. from the Norwegian University of Science and Technology, where he researched control and monitoring solutions for multiphase flows in wells and pipelines. Storkaas can be reached at firstname.lastname@example.org.