Accurate, non-invasive, non-intrusive, and real-time flow metering is required in several applications. For instance, in power generation plants, real-time steam monitoring of the steam quality is required to avoid critical damages that may occur on the blades turbines. In biomedical field, measuring the flow rate of the blood inside the vessels is required to avoid unpredictable strokes which may occur from various cardio-vascular diseases. In oil and gas plants, it is required to ensure proper oil reservoir management and quality monitoring of the fuel being produced. This has led many researchers and companies to suggest and build various kinds of multiphase flow meters. For instance, tomography systems which consist to build two dimensional (2D) and three dimensional (3D) images of the multiphase flow were extensively investigated since they offer the possibility not only to accurately measure the multiphase flow but also to visualize the type of actual flow regime which helps to assess the quality of the design of the pipeline network under the actual flow conditions. Electrical Capacitance Tomography (ECT), Magnetic Induction Tomography (MIT) , and Electrical Resistance Tomography (ERT or EIT) systems are the most used systems used for multiphase flow measurement. They mainly consist to surround across a given section of the probe an array of sensors (i.e. copper electrodes for ECT and EIT and coils for MIT) which are excited in a time multiplexed manner using a predefined sequence and to collect electrical signals the features of which depend on the phases’ distribution within the probe. In spite of the good progress achieved in both the hardware and algorithmic aspects, tomography systems have the problem to not properly cope with the boundaries between phases and also to require an excessive computation time. Modular systolic VLSI architectures using advanced VLSI chips (e.g. FPGA) were suggested to achieve a 2D ECT image reconstruction at a throughput that exceeds 1,000 frames/ second using Linear Back Propagation (LBP) algorithm with eight (8) electrodes. While the availability of highly integrated VLSI chips allowed a substantial reduction of the computation speed, the quality of the output image is still affected by the smoothness constraint of the tomography algorithms. Another limitation of these systems is their disability to cope with small size phases, which is the case of wet gas flow where the droplets of water are dispersed and can be of sub mm-order size. Other concepts for multiphase flow measurement were also considered, such as the ones which allow to “see through” the walls of the conduit using for instance ultrasonic phased array sensors. This has been applied for both blood flow measurement and oil-water and gas flow measurement. The device consists of an array of emitting ultrasonic sensors which are placed on one side of the pipe to transmit ultrasonic waves towards the other diametrically opposite side, which consists of another array of ultrasonic receivers, in a time multiplexed manner. An adequate analysis of the received signals allows to determine the multiphase flow composition as well as to reconstruct its corresponding 2D or 3D image. While this technique is accurate for multiphase flows composed of liquids only (i.e. oil and water), it is impractical for multiphase flows entrained with gas phase. Near-Infrared (NIR)-based devices were recently suggested to measure the flow of moving solid particles (e.g. black powder which are fine particles with high fraction of iron oxide and other chemical contaminants) across a conduit. While this technology has demonstrated its capability to detect accurately very small concentrations of solid contaminants with sub-mm size, it has not been assessed for oil-water-gas multiphase flow fluids. Another technique using gamma-rays was successfully used for multiphase flow measurements. However, in addition of being hazardous, gamma-rays-based probes do not perform well in case of high gas void fraction (i.e. GVF greater than 95%). In the recent years, new emerging techniques based on Nuclear Magnetic Resonance (NMR), Magnetic Resonance Imaging (MRI), and TeraHertz Imaging (THz) were suggested and claimed to be successfully tested for some flow conditions.
The purpose of this tutorial is to present state of the art multiphase flow measurement technologies used so far in oil and gas industries.
Mahmoud Meribout (M’91) received the B.Eng. degree in 1985 and the Ph.D. degree in electronics engineering from the University of Technology of Compiegne, Compiegne, France, in January 1995. He was with Nippon Telegraph and Telephone Corporation, Tokyo, Japan, and NEC Corporation, Tokyo, from 1995 to 2000, where he was involved in several projects related to embedded systems design. In 2008, he joined the Electrical Engineering Department, Petroleum Institute, where he is currently an Associate Professor. His current research interests focusses on embedded systems design.