Research and Realization of Signal Processing Method of Orifice Flow Meter

Abstract: The research and implementation information of the signal processing method of the orifice flowmeter is provided by the excellent flowmeter and flowmeter manufacturers and quotation manufacturers. In practical applications, the output signal of the orifice flowmeter will change slowly and slightly over time. Aiming at this time-varying signal, this paper proposes a recursive algorithm of sliding DTFT (SDTFT) with multi-decimation filter, adaptive lattice notch filter and negative frequency correction. More flowmeter manufacturers select models and price quotations. You are welcome to inquire. The following is the details of the research and implementation of the signal processing method of the orifice flowmeter. In practical applications, the output signal of the orifice flowmeter will change slowly and slightly over time. Aiming at this time-varying signal, this paper proposes to combine a multi-pump filter, an adaptive lattice notch filter and a sliding DTFT (SDTFT) recursive algorithm with negative frequency correction to form a complete set of orifice flowmeter signals. The processing method can not only track the changing frequency and phase, but also have high calculation accuracy when measuring small phases. The entire algorithm requires less computation and no numerical overflow occurs. Developed a signal processing system based on TMS320F28335DSP orifice plate flowmeter, realized a whole set of algorithms, and tested it. Simulation and experimental results show that the method and system developed in this paper are feasible and effective. 1 Introduction The orifice flowmeter can directly measure the mass flow of the fluid with high precision and obtain the fluid density value at the same time. It is one of the most rapidly developing flowmeters. Equations (1) and (2) indicate that the phase and frequency of the signal are changing, and the value at each moment is the value at the previous moment plus a random number, where eΦ(n) and eω(n) are white noise with zero mean, normal distribution, variance 1 and uncorrelated,σΦandσω controls e respectivelyφThe amplitudes of (n) and eω(n) decrease when the signal changes slowly, and increase when the signal changes abruptly.λΦandλΦcontrol separatelyΦThe variation amplitude of (n) and ω(n), the phase variation amplitude should be lower than 1% of the given phase, and the frequency variation should be lower than 0.01% of the vibration frequency, which is more in line with the actual situation. 3 Algorithm principle and derivation 3.1 Multi-decimation filter In order to enhance the noise suppression, the output signal of Coriolis flowmeter is sampled with a higher sampling frequency of 16kHz, and then the multi-decimation filter is used for anti-aliasing filtering and decimation . The multi-decimation filter is divided into two stages [4]. The first stage is 2 decimation and 1, which reduces the actual sampling frequency from 16kHz to 8kHz. The main purpose is to reduce the amount of data. The second stage is 4 to 1, and the sampling frequency is reduced to 2kHz. At the same time, the 30th-order FIR low-pass filter is used, which not only ensures the linear phase, but also in the actual implementation, only the extracted points can be filtered and then extracted, which can reduce the amount of calculation and save time. The coefficients of the multi-decimation filter are obtained by the method of computer-aided design after the cut-off frequency is determined. The simulation results show that the method obtains as much information of the original signal as possible, so the effect is better than that obtained by simply sampling and filtering at 2kHz. 3.2 Adaptive Lattice Notch Filter The parameters of the adaptive notch filter can converge according to the signal characteristics and can estimate the frequency of the signal. The adopted lattice IIR notch filter [1] is formed by cascading two lattice filters of all poles and all zeros. The transfer function is: (3) In order to reduce the computational burden, by fixing the zeros on the unit circle, So that only one parameter can be adjusted to achieve the purpose of adaptive notch. Fix the zero point on the unit circle, even if k1=1, k0 will converge to−cosω, ω is the normalized frequency of the signal,αTo determine the width of the trap, k0 is adaptively adjusted using Burg's algorithm [1]. Since the density of the fluid in the orifice flowmeter is reflected as the change of frequency, it is necessary to track the frequency change of the fluid signal in time. Through a large number of simulation studies, it is found that by adjustingρandλThe final value of , and appropriately increasing the width of the notch filter trap can achieve the tracking of frequency changes while ensuring the accuracy.ρandλThe calculation formulas are shown in formulas (4) and (5). (4) (5) The calculation amount of the lattice-type adaptive notch filter is greatly reduced, and the parameter adjustment is easy. AdjustmentρandλThe final value and the step size of the change can easily track the change of the frequency, and at the same time can achieve a high accuracy. 3.3 SDTFT recursive algorithm and principle of measuring phase difference 3.3.1 SDTFT recursive algorithm The Fourier transform (DTFT) of discrete time series is: (6) DTFT is realized by increasing the calculated sequence length from the first sampling point The calculation of the Fourier coefficients at the specified frequency, which is feasible if the signal is constant over a period of time. However, it cannot be used for time-varying signals. Each sampling point of the time-varying signal contains the new information of the phase change. DTFT confuses and superimposes all the old and new information of the phase change, and cannot reflect the phase change sensitively at all. Therefore, we propose sliding DTFTs to handle time-varying signals. Add a time window of N points to the observed signal. The rectangular window is the simplest time window, and let this time window slide forward as the number of sampling points increases, as shown in Figure 1. With the sliding of the window function, the Fourier transform of the finite-length sequence of N points calculated at each sampling point is the sliding or sliding-window DTFT (SDTFT).

Beijing Sincerity Automatic Equipment Co., Ltd specializes in undertaking corporate offers to cater the needs of different companies.

Applied Materials’ mission is to be the leading supplier of mass flow meter worldwide-through innovation and enhancement of customer productivity with systems and service solutions.

This is especially true when Beijing Sincerity Automatic Equipment Co., Ltd have got a global business that's building bridges between manufacturers and customers across the globe.