Akaike information criterion transit time estimation method based on waveform similarity
By adopting the crossing time estimation method based on the Akagi information criterion with similar waveforms and phase compensation algorithm in the ultrasonic flowmeter, the problems of low measurement sensitivity and noise interference in the small-diameter downhole flowmeter are solved, and higher accuracy and resolution are achieved Rate, suitable for real-time online flow measurement.
Patent Information
- Application Number
- CN202411037929.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-07-31
AI Technical Summary
The existing ultrasonic transit time estimation method is low in measurement sensitivity in small-diameter downhole ultrasonic flowmeters, is susceptible to waveform distortion, spike noise and random noise interference, and has high signal-to-noise ratio requirements, making it difficult to achieve accurate flow measurement.
Using the Akagi Information Criteria (SAIC) method based on waveform similarity, the waveform similarity is calculated by signal cross-correlation method, combined with the FIR bandpass filter and phase compensation algorithm, the accuracy and noise immunity of crossing time are improved, the calculation amount is reduced, and the calculation is enhanced. robustness.
It realizes higher crossing time estimation accuracy and resolution in small-diameter downhole ultrasonic flowmeters, enhances the anti-interference ability to spike noise and random noise, reduces calculation costs, and is suitable for real-time online flow measurement.
Smart Images

Figure CN118981007A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of fluid measurement, and relates to a transit time estimation method based on Akaike information criterion of waveform similarity, which is used for transit time measurement in an ultrasonic flowmeter. Background Art
[0002] An accurate and reliable ultrasonic time of flight (TOF) estimation algorithm is crucial for ultrasonic flowmeters, especially small-diameter downhole ultrasonic flowmeters. The measurement sensitivity of flowmeters in small-diameter pipes is lower than that in large-diameter pipes, which will put higher requirements on the TOF estimation resolution. In addition, the ultrasonic measurement of downhole flowmeters has problems such as waveform distortion, spike noise and harmonic interference, and a highly robust TOF estimation method is required to achieve accurate flow measurement. The various existing TOF estimation methods, including threshold method, window function detection method, cross-correlation method and Akaike Information Criterion (AIC) method, all have their own limitations. For example, the TOF estimation accuracy of the threshold method may be affected by large spike noise and fluctuating signal levels, and it depends on the selection of a suitable threshold coefficient. The robustness and versatility are difficult to meet the actual industrial needs; the window function detection method has the disadvantage of high computational cost and cannot be used for real-time online flow measurement. The accuracy of the cross-correlation method depends on the selection of a suitable reference signal, and its accuracy is easily affected by waveform distortion. Compared with other methods, the AIC method has the advantages of being easy to implement, no need to pre-set parameters, and not being affected by waveform distortion. However, the AIC method has poor noise immunity to random noise and requires a high signal-to-noise ratio of the measurement signal.
[0003] The present invention proposes an AIC TOF estimation algorithm based on waveform similarity. While retaining the advantages of the traditional AIC TOF estimation algorithm such as no need to pre-set parameters and not being affected by waveform distortion, the similarity of two continuous waveforms is used as prior knowledge to further improve the TOF estimation's anti-interference ability against spike noise and random noise. Compared with the neighborhood TOF estimation algorithm based on the AIC framework developed in recent years (X. Qu, "Novel automatic first-arrival picking method for ultrasonic sound-speed tomography," Japanese Journal of Applied Physics, vol. 54, no. 7S1, pp. 07HF10, 2015.) and the spectral entropy TOF estimation algorithm (B. Bühling "Improving onset picking in ultrasonic testing by using a spectral entropy criterion," The Journal of the Acoustical Society of America, vol. 155, no. 1, pp. 544-554, 2024.), the AIC TOF estimation algorithm based on waveform similarity proposed in the present invention simply changes the variance by calculating the covariance of the AIC function, does not require a reference waveform, does not require additional processing steps, has a lower computational complexity, and is more robust to spike noise. Summary of the invention
[0004] The present invention aims to improve the accuracy, noise resistance and resolution of the transit time estimation algorithm in the ultrasonic flowmeter to achieve accurate measurement of the flow rate. To achieve this purpose, the present invention proposes a TOF estimation algorithm based on the similarity-based Akaike Information Criterion (SAIC) of waveform similarity. This method retains the advantages of not requiring pre-setting parameters and not being affected by waveform distortion under the AIC TOF framework. It uses the similarity in the continuous measurement waveform as the algorithm prior information to improve the accuracy and noise resistance, and uses the phase compensation algorithm to break through the sampling frequency limit to improve the measurement resolution. The technical solution of the present invention is as follows:
[0005] A transit time estimation method based on Akaike information criterion of waveform similarity comprises the following steps:
[0006] Step 1: Extract effective ultrasound waveform data
[0007] According to the actual sound path of the ultrasonic flowmeter, the flowmeter range, and the sound velocity of the liquid in the flowmeter, the signal arrival time is estimated; sampling points are intercepted before and after the estimated signal arrival time to form ultrasonic effective waveform data;
[0008] Step 2: Design a FIR bandpass filter to perform zero-phase delay filtering and denoising on the ultrasonic effective waveform data to obtain a zero-phase delay filtered signal;
[0009] Step 3: Estimation of transit time based on Akaike information criterion of waveform similarity;
[0010] The signal cross-correlation method is used to calculate the Akaike information SAIC(k) based on waveform similarity corresponding to the zero-phase-delay filtered signals collected twice by the same ultrasonic sensor;
[0011] The calculated Akaike information based on waveform similarity is exponentially sharpened and the corresponding weight w(k) is calculated:
[0012]
[0013] SAIC min It represents the minimum value of Akaike information SAIC based on waveform similarity;
[0014] Use the weight to locate the minimum value, which is the ultrasonic signal transit time t SAIC :
[0015]
[0016] Step 4: Implement super-resolution transit time correction based on phase compensation.
[0017] In step 1, according to the number of transmitted pulses N p , Flowmeter ultrasonic signal frequency f c , and the ultrasonic measurement sampling frequency f s , respectively intercepted before and after the estimated signal arrival time sampling points constitute the effective ultrasonic waveform data.
[0018] The method of step 2 is: according to the ultrasonic signal frequency f c The ultrasonic measurement sampling frequency f s , design a FIR bandpass filter, and process the ultrasonic effective waveform data in sequence through filtering, time domain flipping, inverse filtering, and secondary flipping to obtain a filtered signal with zero phase delay.
[0019] In step 3, the signal cross-correlation method is used to calculate the formula of the Akaike information SAIC(k) based on waveform similarity corresponding to the zero phase delay filtered signals collected twice by the same ultrasonic sensor:
[0020] SAIC(k)=klog(cov(s1(1,k),s2(1,k)))+Nlog(cov(s1(k+1,N),s2(k+1,N)))
[0021] Where k is the signal sampling point, N is the total length of the signal, and cov(·) represents the signal cross-correlation calculation.
[0022] The method of step 4 is: SAIC The corresponding sampling point k SAIC As the starting point, several sampling points are intercepted backward, and the initial phase θ(k SAIC ):
[0023]
[0024] Where S(K) is the Hilbert transform of the original signal s(k), K fc Indicates the center frequency f c The corresponding Hilbert transform point, imag(·) and real(·) are calculated by taking the real and imaginary parts;
[0025] After obtaining the initial phase, the waveform is corrected and the Akaike information criterion estimation result is calculated to obtain the final transit time t:
[0026]
[0027] The beneficial effects of the present invention are as follows:
[0028] 1. The advantages of the transit time estimation under the Akaike information criterion framework are that it does not require pre-setting parameters and is not affected by waveform distortion.
[0029] 2. Design the Akaike information equation based on the similarity of continuous measurement waveforms to improve the accuracy and noise resistance of transit time measurement;
[0030] 3. Design a super-resolution transit time estimation algorithm based on phase compensation to improve the resolution of transit time measurement. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] The following drawings describe selected embodiments of the present invention, which are exemplary and non-exhaustive or limiting, wherein:
[0032] Figure 1 The ultrasonic flowmeter sound path diagram corresponding to the measurement method of the present invention includes Z-type and π-type ultrasonic flowmeters;
[0033] Figure 2 Ultrasonic signal zero phase delay filtering and denoising flow chart;
[0034] Figure 3The flowchart of transit time estimation based on Akaike information criterion of waveform similarity is shown;
[0035] Figure 4 A flow chart of super-resolution time-of-flight correction based on phase compensation;
[0036] Figure 5 and Figure 6 It is a comparison chart of the effects of different transit time estimation methods, including traditional Akaike information criterion transit time estimation (AIC), weighted Akaike information criterion transit time estimation (WAIC), spectral entropy TOF estimation algorithm (SEC) and the method proposed in this invention (SAIC). DETAILED DESCRIPTION
[0037] The basic scheme of the present invention comprises the following steps:
[0038] Step 1: Extract effective ultrasound waveform data
[0039] According to the actual sound path of the ultrasonic flowmeter, the flowmeter range, and the sound velocity of the liquid in the flowmeter, the signal arrival time is estimated; according to the number of transmitted pulses N p , Flowmeter ultrasonic signal frequency f c , and the ultrasonic measurement sampling frequency f s , respectively intercepted before and after the estimated signal arrival time Sampling points constitute the effective ultrasonic waveform data;
[0040] Step 2: Zero phase delay filtering and denoising of ultrasonic signals;
[0041] According to the ultrasonic signal frequency f c The ultrasonic measurement sampling frequency f s , design a FIR (Finite Impulse Response) bandpass filter H, which filters, flips in time domain, inversely filters, and flips twice the input ultrasonic effective waveform data, and finally outputs a filtered signal with zero phase delay;
[0042] Step 3: Estimation of transit time based on Akaike information criterion of waveform similarity;
[0043] The repetition period of ultrasonic signal transmission and reception in ultrasonic flowmeter is usually 1ms. During this period, the flow velocity change corresponding to the two ultrasonic measurements can be ignored, and the similarity of the measurement waveforms can be used to improve the accuracy of transit time estimation and noise resistance. The filtered signals s1 and s2 with zero phase delay collected by the same ultrasonic sensor twice are used as input, and the corresponding Akaike information SAIC(k) based on waveform similarity is calculated:
[0044] SAIC(k)=klog(cov(s1(1,k),s2(1,k)))+Nlog(cov(s1(k+1,N),s2(k+1,N)))
[0045] Where k is the signal sampling point, N is the total length of the signal, and cov(·) represents the signal cross-correlation calculation;
[0046] The calculated Akaike information SAIC based on waveform similarity is exponentially sharpened and the corresponding weight w(k) is calculated:
[0047]
[0048] SAIC min It represents the minimum value of Akaike information SAIC based on waveform similarity;
[0049] Use the weight to locate the minimum value, which is the ultrasonic signal transit time t SAIC :
[0050]
[0051] Step 4: Achieve super-resolution transit time correction below the sampling interval through phase compensation method.
[0052] Limited by the sampling frequency, the Akaike information criterion transit time has the problem of insufficient time resolution. The super-resolution transit time correction below the sampling interval can be achieved by the phase compensation method. SAIC The corresponding sampling point k SAIC As a starting point, intercept backward sampling points, and calculate the initial phase θ(k SAIC ):
[0053]
[0054] Where S(K) is the Hilbert transform of the original signal s(k), K fc Indicates the center frequency f c The corresponding Hilbert transform point, imag(·) and real(·) are calculated by taking the real and imaginary parts.
[0055] After obtaining the initial phase, the waveform similarity Akaike information criterion estimation result can be corrected and the final transit time t can be calculated:
[0056]
[0057] The embodiments of the present invention are described in detail below with reference to the specification and the drawings.
[0058] Figure 1The ultrasonic flowmeter sound path diagram corresponding to the measurement method of the present invention includes the traditional Z-type ( Figure 1 .a) and π type flow meter ( Figure 1 .b). During the measurement process, the ultrasonic signal generates a square wave pulse signal with a repetition frequency of 1KHz and a center frequency of 1MHz and 20 cycles. During the excitation, another receiving sensor starts to collect the ultrasonic signal and records the waveform and uploads it to the host computer.
[0059] Figure 2 Ultrasonic signal zero phase delay filtering denoising flow chart, here using Z transform as a tool for explanation. First, according to the center frequency and sampling frequency of the ultrasonic measurement signal, a bandpass filter is designed, whose transfer function Z transform is H(Z), and the measurement signal x(t) is bandpass filtered. Define X(z) as the Z transform of x(t), and its filtering result is X(z)H(Z). The time domain flip signal can be obtained by X(z -1 )H(Z -1 ). Bandpass filter the flipped signal again to get X(z -1 )H(Z -1 )H(Z). Now the signal is inverted again in the time domain, and the result is X(z)H(Z -1 )H(Z), if the gain of the bandpass filter itself can be ensured to be |H(z)|=1, the phase changes caused by H(Z) in the filtering result can offset each other, thus achieving ultrasonic signal filtering and denoising with zero phase delay.
[0060] Figure 3 The flowchart of transit time estimation based on Akaike information criterion of waveform similarity;
[0061] The repetition period of ultrasonic signal transmission and reception in ultrasonic flowmeter is usually 1ms. During this period, the change in flow velocity between the two ultrasonic measurements can be ignored, and the similarity of the measurement waveforms can be used to improve the accuracy of transit time estimation and noise resistance. The two received signals of the same ultrasonic sensor are used as input to calculate the corresponding Akaike information equation based on waveform similarity:
[0062] SAIC(k)=klog(cov(s1(1,k),s2(1,k)))+Nlog(cov(s1(k+1,N),s2(k+1,N)))
[0063] Where k is the signal sampling point, N is the total length of the signal, and cov(·) represents the signal cross-correlation calculation.
[0064] The calculated waveform is similar to the Akaike information equation and the corresponding weight of the equation is calculated:
[0065]
[0066] SAICmin Indicates the minimum value of the SAIC equation.
[0067] The equation weight is used to locate the minimum value of the equation, which is the ultrasonic signal transit time:
[0068]
[0069] Figure 4 Flowchart for super-resolution transit time correction based on phase compensation
[0070] The transit time t is calculated by Akaike information criterion SAIC The corresponding sampling point k SAIC As a starting point, intercept backward sampling points, and calculate the initial phase θ(t SAIC ):
[0071]
[0072] Where S(K) is the Hilbert transform of the original signal s(k), K fc Indicates the center frequency f c The corresponding Hilbert transform point, imag(·) and real(·) are calculated by taking the real and imaginary parts.
[0073] After obtaining the initial phase, the waveform similarity Akaike Information Criterion estimation result can be corrected and the final transit time can be calculated:
[0074]
[0075] Figure 5 This is a comparison chart of the effects of different transit time estimation methods. The results in the figure show that when there is spike noise and additive random noise in the ultrasonic measurement signal, the transit time detection results of the AIC, WAIC and SEC methods shift backward and deviate from the actual arrival time, while the SAIC method of the present invention can accurately measure the transit time.
[0076] Since the SEC method is too sensitive to noise, and the difference between the WAIC and AIC methods is small, Figure 6 Only AIC and SAIC methods of the present invention are compared in the long-term detection of . As can be seen from the figure, SAIC method of the present invention has higher accuracy and robustness.
Claims
1. A transit time estimation method based on Akaike information criterion of waveform similarity, comprising the following steps: Step 1: Extract effective ultrasound waveform data Estimate the signal based on the actual sound path of the ultrasonic flowmeter, the flowmeter range, and the sound velocity of the liquid in the flowmeter. Arrival time; sampling points are intercepted before and after the estimated signal arrival time to form effective ultrasonic waveform data; Step 2: Design a FIR bandpass filter to perform zero-phase delay on the ultrasonic effective waveform data. Delayed filtering and denoising to obtain a filtered signal with zero phase delay; Step 3: Estimation of transit time based on Akaike information criterion of waveform similarity; The signal cross-correlation method is used to calculate the zero-phase-delay filtered signal collected twice by the same ultrasonic sensor. The Akaike information SAIC(k) corresponding to the number is based on waveform similarity; The calculated Akaike information based on waveform similarity is exponentially sharpened and the corresponding weight w(k) is calculated: SAIC min It represents the minimum value of Akaike information SAIC based on waveform similarity; Use the weight to locate the minimum value, which is the ultrasonic signal transit time t SAIC : Step 4: Implement super-resolution transit time correction based on phase compensation.
2. The Akaike information criterion transit time estimation method according to claim 1, characterized in that: In step 1, according to the number of transmitted pulses N p , Flowmeter ultrasonic signal frequency f c , and the ultrasonic measurement sampling frequency f s , respectively intercepted before and after the estimated signal arrival time sampling points constitute the effective ultrasonic waveform data.
3. The Akaike information criterion transit time estimation method according to claim 1, characterized in that: The method of step 2 is: according to the ultrasonic signal frequency f c The ultrasonic measurement sampling frequency f s , design FIR bandpass filter The ultrasonic effective waveform data is processed by filtering, time domain flipping, inverse filtering, and secondary flipping in sequence. Obtain a filtered signal with zero phase delay.
4. The Akaike information criterion transit time estimation method according to claim 1, characterized in that: In step 3, the signal cross-correlation method is used to calculate the formula of the Akaike information SAIC(k) based on waveform similarity corresponding to the zero phase delay filtered signals collected twice by the same ultrasonic sensor: 1)SAIC(k)=klog(cov(s1(1,k),s2(1,k)))+Nlog(cov(s1(k+1,N),s2(k)))where k is the signal sampling point, N is the total length of the signal, and cov(·) represents the signal cross-correlation calculation.
5. The Akaike information criterion transit time estimation method according to claim 1, characterized in that: The method of step 4 is: SAIC The corresponding sampling point k SAIC As the starting point, several sampling points are intercepted backward, and the initial phase θ(k SAIC ): Where S(K) is the Hilbert transform of the original signal s(k), K fc Indicates the center frequency f c The corresponding Hilbert transform point, imag(·) and real(·) are calculated by taking the real and imaginary parts; After obtaining the initial phase, the waveform is corrected and the Akaike information criterion estimation result is calculated to obtain the final transit time t:
Citation Information
Patent Citations
Ultrasonic detection method for interface of media using variation information of acoustic impedance
CN110108797A
High-precision estuary short-term tidal level forecasting method
CN110186533A
Ultrasonic wave flight time estimation method
CN113946795A
Signal model establishment and measurement accuracy evaluation of permanent magnet type sodium flowmeter with embedded vortex generator
CN115900898A
Ultrasonic transit time combined measurement method based on threshold value method and cross correlation
CN116735908A
Cited By
An ultrasonic time-of-flight measurement method and apparatus based on echo zero-crossing point
CN122670945A
An ultrasonic time-of-flight measurement method and apparatus based on echo zero-crossing
CN122670945B
