Verarbeitung modulierter Ultraschallsignale in Ein- und Mehrpfadanordnungen bei der korrelativen Durchflussmessung (Processing of Modulated Ultrasonic Signals in One- and Multipath Arrangements for Correlation Flow Measurement)

Abstract

Dieser Artikel beschreibt die Verbesserung des kreuzkorrelativen Ultraschalldurchflussmessverfahrens, sodass im Vergleich mit anderen Ultraschallmessverfahren bei deutlich geringerem Aufwand åhnliche Ergebnisse erreicht werden. Dazu wird zunåchst das Verfahren durch die komplexe Demodulation erweitert. Die Ursachen fçr die Modulationen werden dann in einer Simulation beståtigt und zur Erklårung des Messergebnisses verwendet. Auf diese Weise ist es maeglich, das physikalische Messobjekt zu bestimmen, dessen Messgraeûe mit Hilfe der Kreuzkorrelation ermittelt wird. Abschlieûend lassen sich diese Ergebnisse in Mehrpfadmessungen auf eine Tomographie als Echtzeitvisualisierung des Straemungsprofils anwenden. This paper reports improvements of ultrasonic cross-correlation flow measurements. Through these improvements results similar to those from other ultrasonic measurements are possible but with a significantly reduced technical effort. At first the ultrasonic device is extended by complex demodulation. Simulations are used to confirm and explain the results. This allows the determination of physical objects by cross-correlation. Finally the results are used for multi-path tomography as a real-time visualisation of the flow profile.

Quantitative analysis of the cross-correlation ultrasonic flow meter by means of system theory

“…at lower values of time, and the resulting velocity must be higher than the values determined by using the phase-modulated signals. This is in agreement with the experimental findings of Skwarek [15] who reports, that the velocity values ūcorr / ūb are about 3% higher than those found by using the phase-modulated signals. This observation is validated quantitatively with our theoretical model, as can be seen in figure 16.…”

“…In the beginning it was mentioned that the interaction of the ultrasound wave with the turbulent flow can also result in an amplitude modulation of the wave. The form of the cross-correlation function depends on whether the phase modulation or the amplitude modulation is used for the signal generation [5,15]. The auto-correlation φ 11 (t) obtained from the amplitude-modulated signals is narrower than the φ 11 (t) derived from phase-modulated signals as shown in figure 15 for a special set of parameters (Re = 10 5 ).…”

“…This observation is validated quantitatively with our theoretical model, as can be seen in figure 16. A reason for not using the amplitude-modulated signal is that these signals are more distorted than those obtained from phase modulation (see figure 17 and [15,16]).…”

“…For performing the cross-correlation we use the phase-modulated signals. Details of the generation of the ultrasonic waves and the signal processing are described by Skwarek [15], and for the performance as well as results of the cross-correlation process we refer to Schneider [16].…”

“…Figure 17.Non-dimensional velocity determined by cross-correlation in turbulent pipe flow using either the phase-modulated or the amplitude-modulated ultrasound signals; ūb = bulk velocity from the reference measurements; data from Skwarek[15].…”