Flow measurement of pipeline fluids based on wavenumber-frequency spectrum

Abstract

Flow is a crucial parameter in industrial production, but traditional flowmeters often face limitations due to the field measurement conditions. This paper explores the array signal processing technology used in passive sonar and details how to measure the flow rate in a pipeline through wavenumber-frequency ($k-\omega$) spectrum analysis. Firstly, based on the “Taylor frozen” concept of turbulence, the feasibility of using convective velocity to measure the flow rate in a pipeline is studied, and the composition of the disturbance signal is modeled theoretically. Secondly, the signal amplitude and frequency characteristics of water and air are analyzed as the flow rate changes. The Minimum Variance Distortionless Response (MVDR) beamforming algorithm is utilized to address the “convective ridge,” and the slant stack ($\tau -v$) algorithm helps identify the accurate velocity corresponding to the slope of ridge. Finally, Field experiments were performed using DN50 and DN80 pipelines equipped with piezoelectric sensors. The results showed that the relative errors of water flow and air flow were both less than ±1.0 %, which proved the effectiveness of wavenumber-frequency spectrum method for measuring flow. Therefore, it is of great engineering guiding significance to utilize passive sonar technique to measure pipeline flow.