Validating the design optimisation of ultrasonic flow meters using computational fluid dynamics and surrogate modelling

Domestic ultrasonic flow meters with an intrusive two-stand configuration present a complex flow behaviour due to their unique geometry, which offers an interesting case to evaluate optimisation methods in wall-bounded turbulent flows. In this study, the design and analysis of computer models by computational fluid dynamics is used to predict the turbulent flow and to perform robust design optimisation of the flow meter. The optimisation is accomplished by surrogate modelling based on Kriging, Latin hypercube sampling, and Bayesian strategies to ensure a high-quality and space-filled response surface. A novel function to quantify flow meter measurement uncertainty is defined and evaluated together with pressure drop in order to define the multi-objective optimisation problem. The optimisation Pareto front is shown and compared numerically and experimentally against pressure drop and laser Doppler velocimetry experiments, displaying performance gains and geometrical changes in the 3D space. From the various improved designs sampled experimentally, a 4.9% measurement uncertainty reduction and a 37.4% pressure drop reduction have been shown compared to the analysed baseline case. The applied methodology provides a robust and efficient framework to evaluate design changes, improving ultrasonic flow meters and internal-flow problems with similar features.