CN-121994316-A - Echo characteristic point positioning method and device of ultrasonic water meter and storage medium

CN121994316ACN 121994316 ACN121994316 ACN 121994316ACN-121994316-A

Abstract

The application relates to an echo characteristic point positioning method, device and storage medium of an ultrasonic water meter, wherein the method comprises the steps of correcting a reference flight time value; the method comprises the steps of sorting data sequences of a plurality of feature flight times, carrying out multivariate probability coding on the sorted data sequences, calculating an average value of candidate combination probability amplitude, carrying out pulse time sequence coding and competitive feature significance detection on the sorted data sequences, carrying out variation global optimization on the sorted data sequences, screening first candidate feature point combinations from time sequence position combinations, calculating reconstruction matching degree of the first candidate feature point combinations, carrying out physical constraint inspection on the first candidate feature point combinations, and determining echo feature point positioning results of an ultrasonic water meter from the first candidate feature point combinations according to statistical evidence, pulse confidence degree, reconstruction matching degree and physical constraint score. According to the application, the accurate positioning of the echo signal characteristic points is realized, and the problem of inaccurate positioning of the echo signal characteristic points in the prior art is solved.

Inventors

SUN YINGGUANG
ZHAO WEIGUO
Jing Shaokui
SHEN PINGPING
BAI JINCHAO
WANG FEI
WU XIAOJIE
WANG CHENGLI
JIANG YANFU

Assignees

中国计量大学

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. An echo feature point positioning method of an ultrasonic water meter, which is characterized by comprising the following steps: Collecting echo signals of an ultrasonic water meter, and obtaining a reference flight time value according to the echo signals, wherein the echo signals comprise a plurality of data sequences of characteristic flight time; correcting the reference flight time value according to the fluid temperature acquired in real time to obtain an effective reference value; Sequencing the data sequences of the plurality of characteristic flight times to obtain a sequenced data sequence; performing multi-element probability coding on the ordered data sequence, and calculating the average value of candidate combination probability frames to be used as statistical evidence; performing pulse time sequence coding and competitive feature significance detection on the ordered data sequence, and calculating pulse confidence coefficient of each data point serving as a feature point; performing variation global optimization on the ordered data sequence, screening out a first candidate feature point combination from the time sequence position combination, and calculating the reconstruction matching degree of the first candidate feature point combination; Performing physical constraint inspection on the first candidate feature point combination, and calculating a physical constraint score meeting the consistency of the physical distance and the direction; And determining an echo feature point positioning result of the ultrasonic water meter from the first candidate feature point combination according to the statistical evidence, the pulse confidence, the reconstruction matching degree and the physical constraint score.
2. The method for positioning the echo feature point of the ultrasonic water meter according to claim 1, wherein the step of correcting the reference time-of-flight value according to the fluid temperature acquired in real time to obtain an effective reference value comprises the steps of calculating a temperature compensation factor according to the real-time water temperature through a preset sound velocity temperature model, and dynamically compensating the reference time-of-flight value to obtain the effective reference value at the current temperature.
3. The method for locating echo feature points of an ultrasonic water meter according to claim 1, wherein the step of performing multivariate probability coding on the ordered data sequence to calculate an average value of candidate combined probability frames as statistical evidence comprises calculating probability frames of the data belonging to each preset feature point or non-feature point by using a Gaussian similarity function according to the approach degree of each data to an effective reference value after temperature compensation, and calculating the average value of the candidate combined probability frames as statistical evidence.
4. The method for locating echo feature points of an ultrasonic water meter according to claim 1, wherein the step of performing pulse time sequence coding and competitive feature significance detection on the ordered data sequence and calculating pulse confidence of each data point as a feature point comprises converting the ordered data sequence into a pulse issuing time sequence, introducing a local competition mechanism and a time sequence dependent plasticity rule, and calculating an average value of candidate combined pulse confidence as pulse evidence.
5. The method for positioning echo feature points of an ultrasonic water meter according to claim 1, wherein the step of performing a variational global optimization on the ordered data sequence, screening a first candidate feature point combination from a time sequence position combination, and calculating a reconstruction matching degree thereof comprises the steps of constructing a comprehensive loss function comprising a reconstruction loss, a distance constraint loss and a direction consistency loss, combining a result of global search and probability sampling of the variational optimization, screening the first candidate feature point combination from a feature point combination space, and calculating a direct matching degree with an effective reference value as the reconstruction matching degree.
6. The method for positioning echo feature points of an ultrasonic water meter according to claim 1, wherein the step of performing physical constraint inspection on the first candidate feature point combination and calculating a physical constraint score satisfying the consistency of physical distance and direction includes calculating a difference between an actual distance and an expected distance between adjacent feature points in the first candidate feature point combination, judging the consistency of physical direction of feature point changes, calculating a distance difference and a direction consistency score, and obtaining an overall confidence degree satisfying the physical constraint as the physical constraint score.
7. The method for positioning the echo feature points of the ultrasonic water meter according to claim 1, wherein the step of determining the echo feature point positioning result of the ultrasonic water meter from the first candidate feature point combination according to the statistical evidence, the pulse confidence, the reconstruction matching degree and the physical constraint score comprises the steps of carrying out weighted fusion on the statistical evidence, the pulse confidence, the reconstruction matching degree and the physical constraint score of the first candidate feature point combination, and determining the echo feature point positioning result of the ultrasonic water meter according to the weighted fusion result and the numerical constraint result.
8. The method for positioning the echo feature points of the ultrasonic water meter according to claim 7 is characterized in that the method for positioning the echo feature points of the ultrasonic water meter is determined according to a weighted fusion result and a numerical constraint result and comprises the steps of sequencing the first candidate feature point combination according to the weighted fusion result to obtain a second candidate feature point combination, performing numerical constraint on the second candidate feature point combination, selecting a feature point combination with highest fusion confidence degree meeting the numerical constraint in the second candidate feature point combination as a target feature point combination, selecting the flight time of a fourth wave zero crossing point of the target feature point combination as a feature point, and outputting the positioning result of a current flight time data sequence.
9. An echo feature point positioning device of an ultrasonic water meter, the device comprising: The system comprises an acquisition module, a reference flight time acquisition module and a control module, wherein the acquisition module is used for acquiring echo signals of an ultrasonic water meter and acquiring a reference flight time value according to the echo signals; the correction module is used for correcting the reference flight time value according to the fluid temperature acquired in real time to obtain an effective reference value; the sequencing module is used for sequencing the data sequences of the characteristic flight time to obtain a sequenced data sequence; the coding module is used for performing multi-element probability coding on the ordered data sequence, and calculating the average value of candidate combination probability frames to be used as statistical evidence; The detection module is used for carrying out pulse time sequence coding and competitive feature significance detection on the ordered data sequence, and calculating pulse confidence coefficient of each data point serving as a feature point; The reconstruction matching module is used for performing variation global optimization on the ordered data sequences, screening out first candidate feature point combinations from time sequence position combinations, and calculating the reconstruction matching degree of the first candidate feature point combinations; the physical constraint module is used for carrying out physical constraint inspection on the first candidate feature point combination and calculating a physical constraint score meeting the consistency of the physical distance and the direction; And the determining module is used for determining an echo feature point positioning result of the ultrasonic water meter from the first candidate feature point combination according to the statistical evidence, the pulse confidence coefficient, the reconstruction matching degree and the physical constraint score.
10. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor realizes the steps of the echo feature point positioning method of an ultrasonic water meter according to any one of claims 1 to 8.

Description

Echo characteristic point positioning method and device of ultrasonic water meter and storage medium Technical Field The application relates to the technical field of flow detection, in particular to an echo characteristic point positioning method, an echo characteristic point positioning device and a storage medium of an ultrasonic water meter. Background The ultrasonic water meter is characterized by that its measuring principle is a pair of ultrasonic transducers set on the measuring tube section, alternatively transmitting and receiving ultrasonic waves, and the propagation time difference of ultrasonic waves in forward and backward flow paths can be used for calculating fluid flow speed and flow quantity. The ultrasonic time difference method is widely applied due to the characteristics of high response speed, easy realization of circuit design and the like. The accuracy of the ultrasonic water meter based on time difference measurement is highly dependent on the accurate positioning of the characteristic points of the echo signals. The change in fluid temperature causes changes in fluid density, viscosity, and acoustic impedance, thereby unevenly affecting the propagation attenuation coefficient and beam shape of the ultrasonic wave. This results in systematic scaling and distortion of the overall amplitude, waveform envelope of the echo signal, such that the amplitude dependent on the conventional double-threshold method occurs, resulting in an overall drift of the echo time of flight. Due to the fluctuation of the echo signal and the instantaneous complexity of the flow field, turbulence, bubbles and other disturbances in the complex flow state can cause the instantaneous distortion and timing jitter of the echo signal, and signal attenuation and measurement errors can also occur randomly. Meanwhile, factors such as aging of the transducer, asymmetry of installation and the like introduce long-term measurement deviation. The combined action of these factors causes uncertainty in the measurement of the flight time, and finally affects the measurement accuracy, stability and reliability of the ultrasonic water meter under different working conditions. Aiming at the problems, the traditional time difference method relies on a natural echo zero crossing point, has fuzzy characteristics, is easy to be interfered by noise and has large temperature drift, and the double-threshold method suppresses noise, but the threshold setting relies on experience and has poor adaptability. There is a problem that the positioning of the echo signal characteristic points is inaccurate. Aiming at the problem of inaccurate positioning of echo signal characteristic points in the related technology, no effective solution is proposed at present. Disclosure of Invention In this embodiment, a method, an apparatus and a storage medium for locating echo feature points of an ultrasonic water meter are provided, so as to solve the problem of inaccurate locating of echo signal feature points in the related art. In a first aspect, in this embodiment, there is provided a method for positioning an echo feature point of an ultrasonic water meter, the method including: Collecting echo signals of an ultrasonic water meter, and obtaining a reference flight time value according to the echo signals, wherein the echo signals comprise a plurality of data sequences of characteristic flight time; correcting the reference flight time value according to the fluid temperature acquired in real time to obtain an effective reference value; Sequencing the data sequences of the plurality of characteristic flight times to obtain a sequenced data sequence; performing multi-element probability coding on the ordered data sequence, and calculating the average value of candidate combination probability frames to be used as statistical evidence; performing pulse time sequence coding and competitive feature significance detection on the ordered data sequence, and calculating pulse confidence coefficient of each data point serving as a feature point; performing variation global optimization on the ordered data sequence, screening out a first candidate feature point combination from the time sequence position combination, and calculating the reconstruction matching degree of the first candidate feature point combination; Performing physical constraint inspection on the first candidate feature point combination, and calculating a physical constraint score meeting the consistency of the physical distance and the direction; And determining an echo feature point positioning result of the ultrasonic water meter from the first candidate feature point combination according to the statistical evidence, the pulse confidence, the reconstruction matching degree and the physical constraint score. In some embodiments, the correcting the reference time-of-flight value according to the fluid temperature acquired in real time to obtain an effective reference value includes calcu