Abstract:
The paper addresses the problem of improving the accuracy of vortex flowmeters by means of high-resolution frequency estimation of vortex shedding under conditions of short observation intervals and noisy signals. Traditional methods based on the Fast Fourier Transform (FFT) face fundamental resolution limitations when analyzing short data segments, which reduces their effectiveness in dynamic measurement regimes. As an alternative, a modified Matrix Pencil Method (MPM) is proposed, belonging to the class of high-resolution parametric techniques. The method models the signal as a sum of complex exponentials and provides robust frequency estimation even at low signal-to-noise ratios. A comparative analysis of MPM and FFT was carried out using both simulated and experimental signals from a vortex flowmeter. It is shown that MPM provides more stable frequency estimates, with the standard deviation reduced by an average of 1.5 times. At the same time, the computational complexity of the method is comparable to or even lower than that of FFT due to the small size of the analysis windows. The results demonstrate the potential of MPM for developing algorithms of automatic metrological reliability control. The method can serve as a basis for self-diagnostic systems and correction of measurement errors caused by flow nonstationarity, vibrations, or two-phase flow conditions.
Keywords:vortex flowmeter, frequency estimation, matrix pencil method, high resolution, signal processing, measurement reliability, two-phase flow, parametric methods.
Fulltext:
PDF file (2179 kB)