Evaluation method of transit time difference for clamp-on ultrasonic flowmeters in two-phase flows

• Published in: Experimental thermal and fluid science 2020-04, Vol.112, p.109957, Article 109957
• Main Authors: Murakawa, Hideki, Ichimura, Shuhei, Sugimoto, Katsumi, Asano, Hitoshi, Umezawa, Shuichi, Sugita, Katsuhiko
• Format: Article
• Language: English
• Subjects: Acoustic attenuation; Acoustic impedance; Acoustic noise; Clamp-on; Downstream effects; Energy management; Flow control; Flow rates; Flow velocity; Flowmeters; Guided waves; High temperature; Industrial plants; Information processing; Mist; Noise; Pipes; Signal processing; Standard deviation; Steam flow; Steam flow rate; Time dependence; Time-of-flight flowmeter; Transducers; Transit time; Two phase flow; Ultrasonic signal propagation; Ultrasonic transducers
• ISSN: 0894-1777; 1879-2286
• DOI: 10.1016/j.expthermflusci.2019.109957

•Standard deviation of the signal was proposed for TOF calculation at lower SNRs.•Guided wave is the dominant effect for reducing SNR of the target signal.•Guided wave is strongly influenced by fluctuation of liquid film.•Success ratio of TOF measurements depends on the flow regime in two-phase flow.•Distribution of the standard deviation can be used for the flow regime estimation. To achieve efficient energy management in industrial plants, it is crucial to measure the steam flow rates at the various points of consumption. Clamp-on time-of-flight ultrasonic flowmeters are useful devices to measure the steam flow rates in existing pipes. However, it is difficult to accurately measure the steam flow rates because of the large acoustic impedance difference between the pipe material and fluid, strong signal attenuation in the fluid, and high temperature. In addition, the steam wetness increases with heat losses and the presence of liquid film and droplets that creates noise in the detected ultrasonic signals, making it difficult to distinguish the target signals from the noisy ones. Hence, a new signal processing method is proposed to determine the transit time difference, particularly at lower signal-to-noise ratios. Two ultrasonic transducers were used to measure the ultrasonic time-of-flight, which varies depending on the flow rate. The transmitted ultrasonic signals were time-dependent due to the generation of guided waves in the pipe wall. The standard deviations of the target signals increased when the flow regime transitioned from stratified to annular mist flow. The guided waves significantly influence the success ratio of determining the transit time difference between the ultrasonic signals transmitted from the upstream and downstream transducers. Based on the results, the use of the standard deviation of the target signals is proposed for estimating the transit time difference. Further, as the standard deviation varied significantly depending on the flow regime, it can be used to identify the flow regime as well.