Dry calibration of electromagnetic flowmeters based on numerical models combining multiple physical phenomena (multiphysics)

Abstract

This paper presents a method for dry calibration of an electromagnetic flowmeter (EMF). This method, which determines the voltage induced in the EMF as conductive liquid flows through a magnetic field, numerically solves a coupled set of multiphysical equations with measured boundary conditions for the magnetic, electric, and flow fields in the measuring pipe of the flowmeter. Specifically, this paper details the formulation of dry calibration and an efficient algorithm (that adaptively minimizes the number of measurements and requires only the normal component of the magnetic flux density as boundary conditions on the pipe surface to reconstruct the magnetic field involved) for computing the sensitivity of EMF. Along with an in-depth discussion on factors that could significantly affect the final precision of a dry calibrated EMF, the effects of flow disturbance on measuring errors have been experimentally studied by installing a baffle at the inflow port of the EMF. Results of the dry calibration on an actual EMF were compared against flow-rig calibration; excellent agreements (within 0.3%) between dry calibration and flow-rig tests verify the multiphysical computation of the fields and the robustness of the method. As requiring no actual flow, the dry calibration is particularly useful for calibrating large-diameter EMFs where conventional flow-rig methods are often costly and difficult to implement.

Magnetic field estimation in measurement dead domain for dry calibration of electromagnetic flowmeter

“…This paper extends our previous studies [5,8] to improve dry calibration of flowmeters that experience a long history of service. Specifically, it introduces an estimation method to cost effectively account for the immeasurable field in the MDD.…”

“…In the above equations, the subscripts ( 1 and 2 ) denote different regions of the boundary surface, and B n specifies the normal component of the magnetic flux density [5]. With a few exceptions (such as the magnetic ground surface or remote far field where can be reasonably approximated and arbitrarily set to zero), the potential is not measurable in practice [7].…”

“…For a given [C] m , constraint (5) implies that the estimated [C] p on the pipe wall must satisfy (6) to ensure that Laplace's equation for the MDD is satisfied:…”

An alternative analysis method for evaluating electromagnetic flowmeter performance

“…Furthermore, it is apparent from the literature that other researchers who are very interested in the sensitivity and accuracy of magnetic flowmeters in varied hydraulic conditions either do not appear to be aware of or are incapable of analyzing magnetic flowmeter performance numerically and thus rely on other correlative methods like those described by Martim et al (2019) and Beck et al (2019) as a substitute. In some cases, researchers appear to be as comfortable with the fluid mechanics involved in solving the flow field as they are in understanding the computations required to produce a voltage output for the magnetic flowmeter being modeled (see Fu et al (2010), Beck et al (2021)). However, this appears to be the exception rather than the norm.…”

Single-magnet rotary flowmeter for liquid metals

“…2 The standard approach is to determine the flow rate by measuring the potential difference induced between a pair of electrodes by a flow of conducting liquid in the magnetic field. 3,4 This approach is now well developed and works reliably for common liquids like water, 5 but not so for liquid metals. Major problem in molten metals, especially at elevated temperatures, is the electrode corrosion and other interfacial effects, which can cause a spurious potential difference between the electrodes.…”