Vortex Flowmeter Signal Processing Algorithms

In the industrial production process, fluid (water, air and oil etc) discharge reflects the equipment's load, efficiency and working condition directly. It is one of important parameters which must be detected by control system during the production process. Vortex flow meter namely Karman vortex flow meter has some advantages such as the simple structure, the reliable performance, the long service life, the wide measure range and so on. Because of these advantages, vortex flow meter has been one of important flow measurement meters, and played important roles in metrologies and measurements.Vortex flow meter is subjected to noise disturb caused by pipe vibrations and fluid turbulence and the practical measuring accuracy cannot meet the qualification of the prescriptive indexes. Especially when its velocity is low, the output signal is faint and always cannot be distinguished with noises. It is hard for flow measurement. This paper has mainly researched how to calculate the frequency of vortex signal accurately when the SNR of the signal coming from vortex street sensor is rather low, and realized that the flow could be measured exactly. This paper has introduced the development of vortex flow meter and the research status of vortex signal processing at home and abroad firstly. In addition, the research significance of vortex signal processing has been elucidated. Secondly, the analysis on the vortex flow meter's structure and basic principle, the basic processing circuit and the feature of vortex's signal has been finished. At last, the method of vortex signal processing which includes wavelet analysis method and FFT spectrum analysis method has been put forward. According to the wavelet analysis method, the vortex signal has been decomposed into every frequency range. Then the vortex signal's frequency ranges have been judged through the relation between the amplitude and the flow (frequency) of vortex signal and the signals have been calculated by FFT. The frequencies of vortex signal have been gotten finally. Based on this, this paper has introduced a relative method of spectrum correction aiming at the error coming from the non-integer-period sampling and the calculation by FFT. The gravity method has been used for the correction of the result has improved the calculating accuracy of frequencies of vortex signal.According to the MATLAB simulation, the new and accurate method to calculate the frequencies of vortex signal whose SNR is lower than 1 has been verified. The program composition of the new method on DSP has been finished and the experimental tests have been carried out finally. The result indicates that this method is perfect for the extraction of vortex signal and removal of noise. The flow signal whose SNR is lower than 1 could be extracted exactly, the vortex flow meter's detection accuracy could be improved and measuring range could be extended.