A transit-time ultrasonic flowmeter infers pipe flow rate from the difference between upstream and downstream acoustic propagation times, and its reading is biased whenever the velocity profile is non-uniform across the acoustic path. Zhang, Guo and Lin (2019) derived closed-form profile-correction factors (0.75 for laminar flow, 2n/(2n+1) with a Reynolds-dependent exponent n for turbulent power-law flow) under the approximation that the sound speed greatly exceeds the axial flow component, verified them experimentally, but left their framework incomplete: the validity domain of the approximations is unquantified, the laminar-turbulent transition regime has no correction model, and the required simultaneous solution of flow rate, Reynolds number and n has no specified algorithm. This problem asks for the completion of that correction framework across all steady flow regimes with quantified accuracy.

Background

A transit-time ultrasonic flowmeter measures volumetric flow rate in a pipe from the difference ΔT\Delta T between the propagation times of an acoustic pulse sent obliquely upstream and downstream through the flowing fluid: the downstream pulse travels with the flow and the upstream pulse against it, so ΔT\Delta T is proportional to the flow velocity averaged along the acoustic path. The instrument's working equation, however, assumes a velocity that is uniform over the pipe cross section, while real pipe flow is not uniform. The nature of the profile is governed by the Reynolds number ReRe (the ratio of inertial to viscous forces in the flow): below roughly 2300 the flow is laminar with a parabolic profile u=um(1−s2)u=u_m(1-s^2), where umu_m is the centerline (maximum) velocity and ss is the radius scaled by the pipe radius RR; above roughly 4000 it is turbulent and well described near the wall by the power-law profile u=um(1−s)1/nu=u_m(1-s)^{1/n}, where the exponent nn increases with ReRe; between these limits lies the transition regime, where no universal profile family exists. Because a diametral acoustic path over-weights the faster central flow, the meter over-reads, and a correction factor (the ratio of the true flow rate to the meter reading) is applied. For laminar flow this factor is the classical constant 0.75 (the path average of the parabolic profile exceeds the cross-sectional mean by 4/3); for turbulent flow it depends on nn and hence on ReRe. Numerical studies had earlier quantified profile- and turbulence-induced uncertainties (Iooss, Lhuillier and Jeanneau 2002), and Liu et al. (2014) measured transition-regime correction factors empirically using particle image velocimetry (PIV, an optical technique that records the velocity field from images of tracer particles) for ReRe between 2000 and 20000, but noted the absence of a calculation method for that regime. Zhang, Guo and Lin (2019) gave the analytic derivation: writing the exact transit-time difference as an integral over the scaled radius involving the parameter K=c/(umcos⁡θ)K=c/(u_m\cos\theta), where cc is the sound speed and θ\theta the angle between the acoustic path and the pipe axis, they reduced it to closed form under the assumptions K2≫(1−s2)2K^{2}\gg(1-s^{2})^{2} for laminar and K2≫(1−s)2/nK^{2}\gg(1-s)^{2/n} for turbulent flow, i.e. whenever the sound speed greatly exceeds the axial flow velocity component. The resulting factors, 0.75 and kt=2n/(2n+1)k_t=2n/(2n+1), agreed with gravimetric (weighing-based) reference measurements to 0.976% and 0.25% average relative error respectively, while the transition-zone data deviated from the laminar constant. Applying the turbulent factor requires the tabulated map n=f(Re)n=f(Re) with Re=2ρQ/(πμR)Re=2\rho Q/(\pi\mu R) and Q=ktQ0Q=k_t Q_0 (their Equations (28)-(30)) to be solved simultaneously for the flow rate QQ, since nn itself depends on the unknown QQ. The authors did not quantify when the K2≫K^{2}\gg approximations hold, model the transition regime, or specify an algorithm for the simultaneous solution. Later work has advanced parts of this agenda: Straka et al. (2025) built a Reynolds-dependent profile correction for a clamp-on meter with Monte Carlo uncertainty propagation, and CFD-based disturbance-compensation studies (Papathanasiou et al. 2022; Alashker et al. 2026) treat installation-induced profile distortions separately.

Problem Statement

Complete the velocity-profile correction framework of Zhang, Guo and Lin (2019) for single-path transit-time ultrasonic flowmeters in circular pipes under steady, fully developed flow, so that accurate flow rates with quantified accuracy can be obtained across all flow regimes. Within this single objective, the requested results are: (1) quantify the domain of validity and error bounds of the closed-form correction factors — the laminar constant 0.75 and the turbulent factor kt=2n/(2n+1)k_t=2n/(2n+1) — which were derived under the approximations K2≫(1−s2)2K^{2}\gg(1-s^{2})^{2} and K2≫(1−s)2/nK^{2}\gg(1-s)^{2/n} with K=c/(umcos⁡θ)K=c/(u_m\cos\theta), as functions of the flow-speed-to-sound-speed ratio um/cu_m/c, the propagation angle θ\theta, and the Reynolds number ReRe, including the regimes of high acoustic Mach number and shallow propagation angle where the approximations may fail; (2) provide a correction-factor model covering the laminar-turbulent transition regime, where the measured correction factor deviates from both the laminar constant and the turbulent power-law value, anchored to the laminar and turbulent limits and validated against experimental data; and (3) specify how to solve simultaneously the coupled system n=f(Re)n=f(Re), Re=2ρQ/(πμR)Re=2\rho Q/(\pi\mu R), Q=ktQ0Q=k_t Q_0 (Equations (28)-(30) of the source) needed to evaluate the corrected flow rate from measured transit-time differences and instrument parameters, given that the tabulated n(Re)n(Re) map makes nn depend on the unknown QQ.

References

  1. Zhang, H.; Guo, C.; Lin, J. Effects of Velocity Profiles on Measuring Accuracy of Transit-Time Ultrasonic Flowmeter. Applied Sciences 2019, 9(8), 1648. DOI: 10.3390/app9081648, https://doi.org/10.3390/app9081648 (source paper)
  2. Straka, M.; Höhne, C.; Koglin, C.; Funck, B.; Eichler, T. Reynolds-Dependent Velocity Profile Correction and Its Uncertainty Demonstrated on an Ultrasonic Clamp-On Meter. Metrology 2025, 5(3), 57. DOI: 10.3390/metrology5030057, https://doi.org/10.3390/metrology5030057
  3. Liu, Z.-G.; Du, G.-S.; Shao, Z.-F.; He, Q.-R.; Zhou, C.-L. Measurement of transitional flow in pipes using ultrasonic flowmeters. Fluid Dynamics Research 2014, 46, 055501. DOI: 10.1088/0169-5983/46/5/055501, https://doi.org/10.1088/0169-5983/46/5/055501
  4. Smith, L.; Dixon, S. A deflected ray model of the transit time difference in ultrasonic flow meters applied to non-ideal flow profiles. Flow Measurement and Instrumentation 2024, 97, 102628. DOI: 10.1016/j.flowmeasinst.2024.102628, https://doi.org/10.1016/j.flowmeasinst.2024.102628
  5. Alsaqoor, S.; Piechota, P.; Alahmer, A.; As'ad, S.; Beithu, N.; Wędrychowicz, W.; Andruszkiewicz, A.; Kotomski, P. Examining Transit-Time Ultrasonic Flowmeter Inaccuracies during Changing Gas Velocity Profiles. Processes 2023, 11(5), 1367. DOI: 10.3390/pr11051367, https://doi.org/10.3390/pr11051367
  6. Alashker, M. H.; Zahran, A. A.; Kassem, F. A.-M. Influence of valve-induced flow disturbances on the measurement accuracy of transit-time ultrasonic flowmeters. Measurement Science and Technology 2026, 37(6), 065301. DOI: 10.1088/1361-6501/ae41db, https://doi.org/10.1088/1361-6501/ae41db
  7. Papathanasiou, P.; Kissling, B.; Berberig, O.; Kumar, V.; Rohner, A.; Bezděk, M. Flow disturbance compensation calculated with flow simulations for ultrasonic clamp-on flowmeters with optimized path arrangement. Flow Measurement and Instrumentation 2022, 85, 102167. DOI: 10.1016/j.flowmeasinst.2022.102167, https://doi.org/10.1016/j.flowmeasinst.2022.102167
  8. Iooss, B.; Lhuillier, C.; Jeanneau, H. Numerical simulation of transit-time ultrasonic flowmeters: uncertainties due to flow profile and fluid turbulence. Ultrasonics 2002, 40(9), 1009-1015. DOI: 10.1016/S0041-624X(02)00387-6, https://doi.org/10.1016/S0041-624X(02)00387-6