CN110598520A

CN110598520A - Speed measuring method for preventing strong common mode noise interference in signal

Info

Publication number: CN110598520A
Application number: CN201910571260.7A
Authority: CN
Inventors: 严勇; 卡迈尔·瑞德; 郑格; 胡永辉
Original assignee: Watson Energy Technology (langfang) Co Ltd
Current assignee: Watson Energy Technology (langfang) Co Ltd
Priority date: 2019-06-28
Filing date: 2019-06-28
Publication date: 2019-12-20
Anticipated expiration: 2039-06-28
Also published as: CN110598520B

Abstract

The invention discloses a speed measuring method for preventing strong common mode noise interference in signals, belonging to the technical field of speed measurement. Firstly, filtering high-frequency noise in a signal by adopting a digital filtering algorithm and smoothing the signal. The signal is then time-domain denoised, suppressing the noise frequency components mixed with the original signal. And calculating the displacement corresponding to the peak value of the optimal correlation function to further obtain the transit time, and calculating the correlation speed through the transit time. The invention improves the performance of the related speed measurement algorithm, enables the related speed measurement algorithm to still realize the effective measurement of the related speed in a severe environment, improves the measurement precision, widens the measurement range and improves the measurement repeatability.

Description

Speed measuring method for preventing strong common mode noise interference in signal

Technical Field

The invention belongs to the technical field of speed measurement, and particularly relates to a speed measurement method for preventing strong common-mode noise interference in signals.

Background

In the past twenty years, correlation algorithms have been widely used for speed measurement in industry. The time of flight is obtained by calculating the peak value of the cross correlation function of the output signals of the two sensors arranged in front of and behind the moving direction of the measured object, and then the moving speed is obtained according to the distance between the two sensors. However, the signal from the sensor is susceptible to various external and internal noises, and under extremely severe conditions, the signal is partially or completely submerged in strong noises, so that the signal quality is remarkably reduced, and the measurement is failed. It is therefore important to eliminate noise interference. In selecting the denoising method, the type of noise mixed with the original signal needs to be considered. The most common noises are mains frequency disturbances, vibrations of the mechanical system and intrinsic random noises. The power frequency interference mainly comes from electromagnetic interference of a power line and incorrect grounding of a signal conditioning circuit. The power line noise is centered at 50Hz or 60Hz, the bandwidth does not exceed 1Hz, and most is shielded by the grounded metal of the sensor. However, the addition of a metal layer covering the electrodes and the signal conditioning circuitry does not completely shield external electrical interference, and conventional notch filters cannot be used to remove residual noise because their frequency components (50Hz or 60Hz) may be within the spectrum of the original signal. The second type of noise is caused by vibrations of the mechanical system. The disturbing frequencies of this type of noise are generally composed of high frequency components caused by the vibrations of the connecting cable, the rotating machine main frequency and its harmonics. The third category is inherently random noise, which approximates white noise with a continuous spectral distribution. Therefore, a preferred correlation velocity measurement method is needed to remove these three common noise interferences and achieve effective measurement of correlation velocity.

Disclosure of Invention

The invention aims to provide a speed measurement method for preventing strong common-mode noise interference in signals, which comprises the following steps:

filtering high-frequency noise in the signal by adopting a digital filtering algorithm;

smoothing the filtered signal;

calculating a correlation function of the smoothed signal;

calculating the displacement corresponding to the peak value of the correlation function;

determining a transit time according to the displacement;

and calculating to obtain the related speed according to the transition time.

The invention has the beneficial effects that: the technology is suitable for any measurement based on the correlation algorithm with strong periodic noise, improves the performance of the correlation velocity measurement algorithm, enables the correlation velocity measurement algorithm to still realize effective measurement of the correlation velocity in a severe environment, improves the measurement precision, widens the measurement range, and improves the measurement repeatability.

Drawings

FIG. 1 shows a filtered signal S_1nAnd S_2nOf (2) (where the signal S is exemplified)_1nAnd S_2nWith strong common mode noise).

FIG. 2 shows the filtered signal S_1nAnd S_2nA cross correlation function graph of (a).

FIG. 3 shows the filtered signal S_1nAnd S_2nPreferably a plot of the correlation function.

Fig. 4 is a flow chart of a velocity measurement method for preventing strong common mode noise interference in a signal.

Detailed Description

The invention is described below with reference to the following figures and examples: a velocity measurement method for preventing strong common mode noise interference in signals comprises the following steps: step 1, filtering high-frequency noise in a signal by adopting a digital filtering algorithm; smoothing the filtered signal; step 2, calculating a correlation function of the smoothed signal; step 3, calculating the displacement corresponding to the peak value of the correlation function; determining a transit time according to the displacement; and 4, calculating to obtain the related speed according to the transit time.

Specifically, step 1 filters the high-frequency noise in the signal by using a digital filtering algorithm, which combines a cut-off frequency method based on fourier transform and a median filter, and smoothes the signal while removing the high-frequency noise component. In one example, a 5 th order IIR butterworth filter is used to suppress noise outside of the signal frequency, and then a median filter with a window size of 50 is used to smooth the signal.

Specifically, step 2, a first noise-free signal S is calculated₁(t) autocorrelation function R_S1S1A first noise-free signal S₁(t) and a second noise-free signal S₂(t) cross-correlation function R_S1S2Can be written as

Where N is the number of samples in the correlation calculation, and m (m 0.., N) is the number of delay points; if the first noiseless signal S₁(t) and a second noise-free signal S₂(t) is disturbed by strong common mode periodic noise n (t) associated therewith, the first noisy signal S_1n(t) autocorrelation function R_S1nS1nIs composed of

While the first noisy signal S_1n(t) and a second noisy signal S_2n(t) cross-correlation function R_S1nS2nIs composed of

Wherein S is_1n(k) And S_2n(k) Are respectively digitized signals S_1n(t) and S_2n(t)。

FIGS. 1 and 2 are respectively filtered signals S_1nAnd S_2nAnd two filtered signals S_1nAnd S_2nThe cross correlation function of (a). Although the use of a digital low-pass filter eliminates high frequency noise interference, the filtered signal is still subject to noise interference associated with the original signal, with no clear dominant peak in the correlation function of the signal.

Since noise is periodic and signal dependent, R is not sufficiently low for signal-to-noise ratio_S1n，R_nS1，R_nS2And R_nnThese correlation terms do not go to zero and the correlation function is disturbed. To address this problem, the present disclosure proposes an optimized correlation function R_modIn which S is_1nAnd (S)_1n-S_2n) Has a correlation function of

S_2nAnd (S)_1n-S_2n) Has a correlation function of

The subtraction of equation (11) and equation (12) yields

Wherein R is_modIs S_1nAnd S_2nBy a noise-free signal S only₁And S₂The autocorrelation and cross-correlation functions of;

in particular, step 3, a preferred correlation function R is calculated_modObtaining the transition time tau by the displacement corresponding to the peak value;

specifically, in step 4, the correlation velocity v is calculated as L/τ, where L is the distance between the two sensors.

FIG. 3 shows two filtered signals S_1nAnd S_2nPreferred correlation function, R_S1S1And R_S2S2At the same time delay, i.e. the period T of the noise. Cross correlation function R_S1S2And R_S2S1Are inverted in equation (13) and are located at the transit times τ and (T- τ), respectively. Preferred correlation function R_modThe influence of periodic noise has been significantly eliminated and the correlation peak becomes more significant, and from the distance L between the two sensors, the correlation velocity v-L/τ can be calculated.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Claims

1. A method for measuring a velocity to prevent strong common mode noise interference in a signal, wherein the method comprises:

smoothing the filtered signal;

calculating a correlation function of the smoothed signal;

determining a transit time according to the displacement;

and calculating to obtain the related speed according to the transition time.

2. The method of claim 1, wherein the digital filtering algorithm is determined according to a fourier transform-based cut-off frequency method and a median filter.

3. The method of claim 1, wherein computing the correlation function of the smoothed signal comprises:

calculating a first noise-free signal S₁(t) autocorrelation function R_S1S1，

Calculating a first noise-free signal S₁(t) and a second noise-free signal S₂(t) cross-correlation function R_S1S2，

Wherein N is the number of samples in the correlation calculation, m is 0.. and N is the number of delay points;

when S is₁(t) and S₂(t) calculating a first noisy signal S when disturbed by strong common mode periodic noise n (t) associated therewith_1n(t) autocorrelation function R_S1nS1n，

Calculating a first noisy signal S_1n(t) and a second noisy signal S_2n(t) cross-correlation function R_S1nS2n，

Wherein S is_1n(k) And S_2n(k) Are each S_1n(t) and S_2n(t) the digitized signal;

determination of S_1nAnd (S)_1n-S_2n) Is related to function R_S1n(S1n-S2n)，

Determination of S_2nAnd (S)_1n-S_2n) Is related to function R_S2n(S1n-S2n)，

Will correlate with function R_S1n(S1n-S2n)And a correlation function R_S2n(S1n-S2n)Subtracting to determine a first noisy signal S_1n(t) and a second noisy signal S_2n(t) optimized correlation function R_mod，

4. The method of claim 1, wherein calculating a correlation velocity from the time of flight comprises:

the correlation velocity v is calculated as L/τ, where L is the spacing of the two sensors and τ is the transit time.