Research On Digital Signal Processing Methods Of Coriolis Mass Flowmeters

Coriolis mass flowmeters are widely used in many various industrial fields due to their following advantages: measuring the mass flow rate directly, high measurement accuracy and measuring density of fluids at the same time, etc. The conventional Coriolis mass flowmeters adopt the signal processing method based on analog circuits, which have many shortcomings, such as being sensitive to noise. So we use digital signal processing methods to deal with the signals of Coriolis meters.This paper mainly studies the signal enhancing and frequency estimating methods based on adaptive line enhance technique and the time interval calculation based on discrete Fourier transform and sliding Goertzel algorithm for Coriolis mass flowmeters.The main work of this disseitation is as follows:(1) According to the characteristics of the output signals, a time-varying signal model based on random walk model is proposed. This model and corresponding signal processing methods are helpful for researches on dynamic characteristics of Coriolis mass meters.(2) Using direct type IIR adaptive line enhancer to enhance Coriolis mass meters signal and estimate its frequency. Different implementations of this type enhancer for time-stationary and time-varying signal are also given.(3) Due to the shortcomings of direct type IIR adaptive line enhancer, using lattice type IIR adaptive line enhancer is proposed. Different implementations of this enhancer for time-stationary and time-varying signal are also studied. During the study for time-varying signal, the original algorithm is improved. The improved algorithm can track frequency variation of the time-varying signal with no extra calculation burden.(4) For the possibility of implementing Coriolis mass flowmeters with FPGA, using normalized lattice type IIR adaptive line enhancer to enhance Coriolis mass flowmeters signals and estimate their frequencies are proposed.(5) Using discrete Fourier transform with Hanning window to calculate the time interval of Coriolis meters signals is studied. Then according to the disadvantages of DFT method, using sliding Goertzel algorithm is proposed. This algorithm can avoid the influence of non-integral period sampling and calculate the time interval in real time.(6) Implementing lattice type IIR adaptive line enhancer and sliding Goertzel algorithm on ADSP21065L DSP. The implement is tested, and the result shows the methods studied in this paper are effective.