Research On Frequency Estimation Methods For Coriolis Mass Flowmeter Based On Adaptive Notch Filter

Coriolis mass flowmeter(CMF) can directly obtain the mass flow and has a lot of excellent charateristics such as high precision, good applicability to different flow. Consequently, as one of the most quickly developed and the most representative mass flow meters at present, CMF has gathered broad attention from petroleum industry, chemical industry and so on.The mass flow rate of CMF is proportional to the time difference of the two signals outputed by its primary instruments. Exactly and timely detecting the frequency and phase difference, and then calculating the time difference is the key of CMF signal processing. Meanwhile, the measurement accuracy of phase difference is depends on the signal frequency tracking estimation accuracy. However, the existing frequency tracking estimation methods for CMF are mainly based on adaptive notch filter(ANF), which is deficient in the convergence speed, frequency tracking accurately for long time, and cannot counterpoise the convergent rate and the tracking precision, so it is unsatisfied to be applied to applications.In order to solve the problems mentioned above, frequency estimation methods for CMF based on adaptive notch filter are studied. The main researches accomplished in the dissertation include:(1) The frequency estimation methods for CMF signals are summarized.In order to overcome the current problems and lay the foundation to propose a new frequency estimation method for CMF, the existing domestic and intenational relevant frequency estimation methods for CMF are analysed, and the advantages and disadvantages of these methods are compared.(2) The signal models of CMF are analyzed.The time invariant and time varying signal models are compared based on the measurement principle and the features of CMF, which contributes to study the dynamic performance of frequency estimation method for CMF.(3) Three kinds of adaptive notch filters' performance are analyzed.In order to understand the advantages and disadvantages of adaptive notch filters intuitively, and propose a new and a better method to solve the problem, the adaptive lattice notch filter(L-ANF), the simplified adaptive lattice notch filter(SL-ANF) and the new adaptive notch filter(New-ANF) are compared through analysis its performance.(4) An improved frequency estimation method for CMF based on simplified adaptive lattice notch filter is proposed.The simplified adaptive lattice notch filter has the characteristics, such as simple structure, strong tracking ability. While it still exists non-absolute convergence problem, which leads to poor persistent estimation precision. In order to overcome the problem, an improved frequency estimation method for CMF based on simplified adaptive lattice notch filter is proposed by adopting the feedback control theory, which is designed to evaluate the frequency tracking quality and increase frequency tracking ability.(5) A signal frequency estimation method for CMF is proposed based on interpolation FFT and improved simplified adaptive lattice notch filter.In order to remove the influence of existing adaptive notch filters on convergence, a new signal frequency estimation method for CMF is proposed based on interpolation FFT and improved simplified adaptive lattice notch filter, which used the interpolation FFT to estimate the signal frequency firstly, and then adopted the improved simplified adaptive lattice notch filter to track the frequency.