ABSTRACT
The vortex flow-meter is widely used in industrial、national defense、scientific
research and nearly all areas of people’s life because of its small pressure losses、wide
range、high accuracy、adaptability of various media and high reliability, etc. However, in
the actual application process, signal from the vortex flower-meter sensor will be mixed
with different forms of noise due to pipe vibration and fluid pulsation effects, making
the measurement inaccurate. In order to solve the problem, we use modern digital signal
processing to approach to signal of vortex flow-meter sensor, developed corresponding
digital signal processing system of vortex flow-meter to improve the accuracy of vortex
flow-meter.
Wavelet transform is known as a mathematical microscope, the ability to
multi-resolution analysis. this paper use MALLAT algorithm of wavelet transform to
decompose the vortex flow sensor signal, capture the flow of information from a
mixture of the vortex signal noise signal, and then estimate the vortex frequency, and
flow meter. Algorithm simulation results show that the algorithm can achieve estimating
the vortex frequency, with strong noise immunity. Calculation is more simple to upgrade
and dual lifting wavelet transform can also be achieved by lifting scheme than direct
MALLAT algorithm, lifting scheme wavelet transform to achieve higher accuracy, this
article also uses the lifting scheme wavelet transform, vortex flow sensor signal
processing. But on lifting scheme wavelet transform not only need to decompose, but
also need the inverse transform for signal reconstruction, to facilitate estimates of the
frequency of vortex.
In this paper, we use signal processing algorithm of wavelet transform to deal
with signal of vortex sensor, and using TI’s TMS320LF2407A developed digital signal
processing system of vortex flow-meter.TMS320LF2407A with powerful computing
capabilities, and abundant-chip resources to meet the real-time processing in the vortex
signal while reducing the system size and cost, improve system reliability. The hardware
of the digital signal processing system include the main processor、the analog signal
input conditioning circuitry、LCD display、EEPROM memory、4~20mA output circuit、
keyboard input circuit、under voltage protection circuit and the pulse output circuit. The
debug results show that the developed system is indeed able to achieve real-time study
of wavelet signal processing algorithms, in order to ensure flow measurement accuracy
of the instrument; but also the instrumentation requirements of other functions such as