Gas ultrasonic flow meter signal processing methods, devices, equipment and media

Abstract

This application relates to the technical field of signal processing, and in particular to a signal processing method, apparatus, device, and medium for ultrasonic gas flow meters. The method involves acquiring an acoustic signal at a target flow point; determining the signal-to-noise ratio (SNR), peak salience, and waveform coherence of the acoustic signal; determining the signal quality index of the acoustic signal based on the SNR, peak salience, and waveform coherence; determining a reconstruction strategy for the acoustic signal based on the signal quality index and a preset decision range; reconstructing the acoustic signal according to the reconstruction strategy to obtain a reconstructed signal; and determining the propagation time based on the reconstructed signal to determine the gas flow rate at the target flow point. This method enhances the anti-interference performance of the time-of-flight ultrasonic gas flow meter signal waveform under complex operating conditions, improves the accuracy of propagation time measurement, and makes the measurement of gas flow rate more precise.

Description

Technical Field

[0001] This application relates to the field of signal processing technology, and in particular to signal processing methods, apparatus, equipment and media for ultrasonic gas flow meters. Background Technology

[0002] Ultrasonic flow meters are instruments that measure flow velocity by detecting the effect of an object's flow on the propagation of ultrasonic waves. Originating in the 1930s, ultrasonic flow meters come in a wide variety of types. According to their measurement principles, they can be divided into the propagation velocity difference method, the Doppler method, the noise method, and the beam deflection method. The time difference method is suitable for fluids that can propagate sound waves, and therefore it is highly valued in the entire flow meter industry. Accurately identifying the arrival time of the ultrasonic received signal is an important issue in the development of time difference ultrasonic flow meters.

[0003] When using the time-of-flight method to measure gas flow, due to the compressibility of gas and the influence of complex operating conditions such as interference noise and turbulence in the ultrasonic operating frequency band, the waveform of the ultrasonic signal actually received by the ultrasonic probe at the receiving end is often distorted. In many cases, this causes a large deviation in the measurement of ultrasonic wave propagation time, resulting in inaccurate vertical flow or large fluctuations, which affects the measurement of gas flow. Summary of the Invention

[0004] This application provides a signal processing method, apparatus, device, and medium for a gas ultrasonic flow meter, which can enhance the anti-interference performance of the signal waveform of the time-difference gas ultrasonic flow meter under complex working conditions, improve the measurement accuracy of propagation time, and make the measurement of gas flow more precise.

[0005] On one hand, embodiments of this application provide a signal processing method for a gas ultrasonic flow meter, the method comprising:

[0006] Acquire the acoustic signal at the target flow point;

[0007] Based on the acoustic signal, the signal-to-noise ratio, peak salience, and waveform coherence of the acoustic signal are determined. The signal-to-noise ratio is used to assess the degree to which the signal is contaminated by noise, the peak salience is used to assess the clarity of the peak, and the waveform coherence is used to assess the symmetry of the flow field.

[0008] The signal quality index of the acoustic signal is determined based on the signal-to-noise ratio, the salience of the main peak, and the waveform coherence.

[0009] Based on the signal quality index and the preset decision range, a reconstruction strategy for the acoustic signal is determined;

[0010] The acoustic signal is reconstructed according to the reconstruction strategy to obtain a reconstructed signal;

[0011] Based on the reconstructed signal, the propagation time is determined to determine the gas flow rate at the target flow point.

[0012] Optionally, before acquiring the acoustic signal of the target flow point, the method further includes:

[0013] The characteristic parameters of each gas ultrasonic flow meter are obtained, including resonant frequency, dynamic resistance and mechanical quality factor;

[0014] Based on the resonant frequency, dynamic resistance, and mechanical quality factor, as well as the corresponding weight values, the comprehensive matching index is determined.

[0015] The gas ultrasonic flow meters whose comprehensive matching index is lower than the preset threshold are matched to obtain a gas ultrasonic flow meter combination.

[0016] The acquisition of the acoustic signal at the target flow point includes:

[0017] Acquire the acoustic signal of the gas ultrasonic flow meter assembly at the target flow point.

[0018] Optionally, the reconstruction strategy includes a first reconstruction strategy, the first reconstruction strategy including a parameterized fine-tuning reconstruction strategy, the reconstruction signal including a first reconstruction signal, and the step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes:

[0019] When the reconstruction strategy is determined to be the first reconstruction strategy, feature parameters are extracted based on the acoustic signal. The feature parameters include the total duration of the whole wave, the center frequency, the rise time of the main peak, and the fall time of the main peak.

[0020] The asymmetry of the main peak is determined based on the rise time and fall time of the main peak, and the asymmetry of the main peak is used to reflect the changes in the flow field.

[0021] Based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, a matching template corresponding to the acoustic signal is selected from a preset feature library;

[0022] The correction error is determined based on the first matching signal in the matching template and the acoustic signal;

[0023] The acoustic signal is reconstructed based on the correction error to obtain a first reconstructed signal.

[0024] Optionally, before selecting a matching template corresponding to the acoustic signal from a preset feature library based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, the method further includes:

[0025] Under preset operating conditions, acquire multiple upstream and downstream acoustic signals from multiple flow points;

[0026] The upstream and downstream acoustic signals are clustered using a preset clustering algorithm to obtain the reference signal for each flow point;

[0027] Feature parameters are extracted from the reference signal to obtain reference global features, reference main peak local features, reference time series features, and reference frequency domain features, as well as the corresponding propagation time values.

[0028] Based on the aforementioned global reference features, local reference main peak features, time-series reference features, and frequency-domain reference features, along with their corresponding propagation time values, a preset feature library is constructed.

[0029] Optionally, the step of selecting a matching template corresponding to the acoustic signal from a preset feature library based on the main peak asymmetry, total duration of the full wave, and intermediate frequency includes:

[0030] Based on the signal quality index and preset adjustment conditions, determine the dynamic weight of each feature parameter;

[0031] A weighted feature vector is constructed based on each feature parameter and its corresponding dynamic weight;

[0032] The similarity between the weighted feature vector and the reference signal vector in the preset feature library is calculated using a preset similarity algorithm.

[0033] Based on the similarity, a matching template corresponding to the sound wave signal is selected.

[0034] Optionally, the reconstruction strategy includes a second reconstruction strategy, the second reconstruction strategy includes a model waveform reconstruction strategy, the reconstructed signal includes a second reconstructed signal, and the step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes:

[0035] When the reconstruction strategy is determined to be the second reconstruction strategy, the acoustic signal and the matching template are aligned at the end to obtain a serial signal;

[0036] The series signal is input into a preset reconstruction model;

[0037] The encoder of the preset reconstruction model is used to capture the context information of the serial signal;

[0038] The decoder of the preset reconstruction model is used to capture local details of the serial signal;

[0039] The second reconstructed signal is fitted based on the context information and the local details.

[0040] Optionally, the reconstruction strategy includes a third reconstruction strategy, the third reconstruction strategy includes a conservative correlation reconstruction strategy, the reconstructed signal includes a third reconstructed signal, and the step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes:

[0041] The acoustic signal is enhanced to obtain an enhanced acoustic signal;

[0042] Perform a Fast Fourier Transform on the enhanced acoustic signal to obtain the signal spectrum;

[0043] Based on the signal spectrum, determine the cross-power spectrum;

[0044] The cross-power spectrum is weighted and subjected to inverse Fourier transform to obtain the generalized cross-correlation function;

[0045] The reconstructed signal is determined based on the generalized cross-correlation function.

[0046] On the other hand, embodiments of this application provide a signal processing device for a gas ultrasonic flow meter, the device comprising:

[0047] The acquisition module is used to acquire the acoustic signal of the target flow point;

[0048] The determination module is used to determine the signal-to-noise ratio, peak salience, and waveform coherence of the acoustic signal based on the acoustic signal. The signal-to-noise ratio is used to assess the degree to which the signal is contaminated by noise, the peak salience is used to assess the clarity of the peak, and the waveform coherence is used to assess the symmetry of the flow field.

[0049] The determining module is further configured to determine the signal quality index of the acoustic signal based on the signal-to-noise ratio, main peak salience, and waveform coherence.

[0050] The determining module is also used to determine a reconstruction strategy for the acoustic signal based on the signal quality index and a preset decision range;

[0051] A signal reconstruction module is used to reconstruct the acoustic signal according to the reconstruction strategy to obtain a reconstructed signal;

[0052] The determining module is also used to determine the propagation time based on the reconstructed signal in order to determine the gas flow rate at the target flow point.

[0053] In another aspect, embodiments of this application provide an electronic device, the device including: a processor and a memory storing computer program instructions;

[0054] When the processor executes the computer program instructions, it implements the gas ultrasonic flow meter signal processing method as described in the first aspect.

[0055] In another aspect, embodiments of this application provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the gas ultrasonic flow meter signal processing method as described in the first aspect.

[0056] The gas ultrasonic flow meter signal processing method, apparatus, device, and computer storage medium of this application embodiment can determine the signal quality index of the acoustic signal by the signal-to-noise ratio, peak saliency, and waveform coherence of the acoustic signal. The signal quality index quantifies the degree of interference of the acoustic signal. Different reconstruction strategies are selected according to different degrees of interference. Then, the propagation time is determined based on the reconstructed signal obtained by different reconstruction strategies. This can enhance the anti-interference performance of the time-difference gas ultrasonic flow meter signal waveform under complex working conditions, improve the measurement accuracy of acoustic propagation time, and thus improve the accuracy of flow detection. Attached Figure Description

[0057] Figure 1 This is a schematic flowchart of a gas ultrasonic flow meter signal processing method provided in an embodiment of this application;

[0058] Figure 2 This is a schematic diagram of the structure of a signal processing device for a gas ultrasonic flow meter provided in an embodiment of this application;

[0059] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0060] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0061] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0062] To address the problems of existing technologies, embodiments of this application provide a signal processing method, apparatus, device, and medium for a gas ultrasonic flow meter. In these embodiments, the signal quality index of the acoustic signal can be determined by its signal-to-noise ratio, peak saliency, and waveform coherence. The signal quality index quantifies the degree of interference affecting the acoustic signal. Different reconstruction strategies are selected based on the degree of interference. Then, the propagation time is determined based on the reconstructed signals obtained from these strategies. This enhances the anti-interference performance of the time-of-flight gas ultrasonic flow meter signal waveform under complex operating conditions, improves the accuracy of acoustic propagation time measurement, and thus improves the accuracy of flow detection.

[0063] The following section first introduces the signal processing method for the ultrasonic gas flow meter provided in the embodiments of this application.

[0064] Figure 1 A schematic flowchart of a gas ultrasonic flowmeter signal processing method according to an embodiment of this application is shown. Figure 1 As shown, the signal processing method for the gas ultrasonic flow meter may include S101-S106;

[0065] S101, acquire the acoustic signal of the target flow point.

[0066] In this embodiment of the application, for a gas transmission pipeline, multiple flow points are usually set for gas flow detection. When detecting a section of the pipeline, different ultrasonic gas flow meters can be used. In this process, the calibrated ultrasonic gas flow meters can be used to obtain the data. Each calibrated ultrasonic gas flow meter has a corresponding matching combination, which can ensure the accuracy of the measured propagation time.

[0067] S102, based on the acoustic signal, determine the signal-to-noise ratio, main peak salience, and waveform coherence of the acoustic signal.

[0068] In this embodiment, the signal-to-noise ratio (SNR) is used to assess the degree of noise contamination of the signal, the salience of the main peak is used to assess the clarity of the main peak, and the waveform coherence is used to assess the symmetry of the flow field. The signal quality of the acoustic signal can be analyzed through the above-mentioned SNR, salience of the main peak, and waveform coherence, providing data support for determining different reconstruction strategies in the future, so as to more accurately calculate the propagation time and gas flow rate.

[0069] S103. Determine the signal quality index of the acoustic signal based on the signal-to-noise ratio, the salience of the main peak, and the waveform coherence.

[0070] In this embodiment, signal quality can be determined by signal-to-noise ratio, peak saliency, and waveform correlation to quantify the signal quality. The signal quality index can be a normalized value of 0-1. When the signal quality index is greater than a first preset threshold, it indicates that the signal quality is high and can be fine-tuned. When the signal quality index is greater than a second preset threshold but less than or equal to the first preset threshold, it indicates that the acoustic signal is poisoned and distorted and needs to be enhanced. When the signal quality index is less than or equal to the second preset threshold, it indicates that the signal quality is too poor and can be marked as invalid or conservatively estimated.

[0071] S104 determines the reconstruction strategy for the acoustic signal based on the signal quality index and the preset decision range.

[0072] In this embodiment, different reconstruction strategies can be adopted for different signal quality indices to obtain more accurate propagation time. The preset decision range includes greater than a first preset threshold, greater than a second preset threshold but less than or equal to a second preset threshold, and less than or equal to a second preset threshold.

[0073] S105, The acoustic signal is reconstructed according to the reconstruction strategy to obtain the reconstructed signal.

[0074] In the embodiments of this application, different reconstruction strategies are used for signal reconstruction based on different signal qualities. The reconstructed signals obtained are all consistent with reality, ensuring the accuracy of propagation time and enabling more accurate calculation of gas flow rate.

[0075] S106, based on the reconstructed signal, determines the propagation time in order to determine the gas flow rate at the target flow point.

[0076] In this embodiment, the signal quality index of the acoustic signal can be determined by the signal-to-noise ratio, peak saliency, and waveform coherence. The interference level of the acoustic signal can be quantified by the signal quality index. Different reconstruction strategies can be selected according to the different levels of interference. Then, the propagation time can be determined based on the reconstructed signals obtained by different reconstruction strategies. This can enhance the anti-interference performance of the time-difference gas ultrasonic flow meter signal waveform under complex working conditions, improve the measurement accuracy of acoustic propagation time, and thus improve the accuracy of flow detection.

[0077] In some other embodiments, in order to obtain a more accurate acoustic signal, the method may further include, before S101:

[0078] Obtain the characteristic parameters of each gas ultrasonic flow meter;

[0079] The overall matching index is determined based on the resonant frequency, dynamic resistance, and mechanical quality factor, as well as their corresponding weight values.

[0080] Match the gas ultrasonic flow meters whose comprehensive matching index is lower than the preset threshold to obtain a gas ultrasonic flow meter combination.

[0081] S101 can specifically include:

[0082] Acquire the acoustic signal of the gas ultrasonic flow meter combination at the target flow point.

[0083] In this embodiment, to ensure that each ultrasonic gas flow meter operates under the same conditions and that the measured results are unaffected by physical factors, matching can be performed based on the characteristic parameters of each ultrasonic gas flow meter. These characteristic parameters may include resonant frequency, dynamic resistance, and mechanical quality factor. Specifically, the resonant frequency is the frequency characteristic of the transducer. The dynamic resistance is the resistance value exhibited by the transducer when operating at the resonant frequency, which is also the lowest point of the transducer's impedance characteristic curve. The mechanical quality factor describes the mechanical loss of the transducer in the resonant state and reflects the steepness of the transducer's impedance characteristic curve. During matching, the relative error of the resonant frequency is required to be less than 0.3%, the relative error of the dynamic resistance less than 10%, and the relative error of the mechanical quality factor less than 5%.

[0084] In this embodiment of the application, the comprehensive matching index can be determined by the following formula:

[0085]

[0086] Where M is the comprehensive matching degree index, w1, w2, and w3 are the corresponding weight values, Fs is the resonant frequency, R1 is the dynamic resistance, and Qm is the mechanical quality factor. The change in the resonant frequency of the two gas ultrasonic flow meters. 1 represents the change in dynamic resistance between the two ultrasonic gas flow meters. The values ​​represent the changes in the mechanical quality factor of the two gas ultrasonic flow meters.

[0087] During the matching process, a preset threshold can be set for matching. That is, the two gas ultrasonic flow meters can only be matched when the comprehensive matching index is lower than the preset threshold, so as to ensure the scientific and optimal matching. As an example, the preset threshold can be 0.02.

[0088] After obtaining the gas ultrasonic flow meter assembly, in step S101, the acoustic signal of the gas ultrasonic flow meter assembly can be acquired to ensure that the physical interference received by the acoustic signal is consistent.

[0089] In some other embodiments, in S103, the signal quality index can be obtained by weighted averaging and summing the signal-to-noise ratio, the main peak significance, and the waveform correlation, and then normalizing the summation.

[0090] In S104 and S105, the reconstruction strategy may include a first reconstruction strategy, a second reconstruction strategy, and a third reconstruction strategy. The first reconstruction strategy may be a parameterized fine-tuning reconstruction strategy, the second reconstruction strategy may be a model waveform reconstruction strategy, and the third reconstruction strategy may be a conservative correlation reconstruction strategy. The reconstructed signal may include a first reconstructed signal reconstructed by the first reconstruction strategy, a second reconstructed signal reconstructed by the second reconstruction strategy, and a third reconstructed signal reconstructed by the third reconstruction strategy.

[0091] Specifically, in S104, when the signal quality index is greater than the first preset threshold, the reconstruction strategy can be determined as the first reconstruction strategy; when the signal quality index is greater than the second preset threshold and less than or equal to the first preset threshold, the reconstruction strategy can be determined as the second reconstruction strategy; and when the signal quality index is less than or equal to the second preset threshold, the reconstruction strategy can be determined as the third reconstruction strategy. As an example, the first preset threshold can be 0.7 and the second preset threshold can be 0.3.

[0092] When the reconstruction strategy is the first reconstruction strategy, S105 may include:

[0093] Extract feature parameters based on the acoustic signal;

[0094] The asymmetry of the main peak is determined based on the rising and falling times of the main peak.

[0095] Based on the asymmetry of the main peak, the total duration of the whole wave, and the intermediate frequency, a matching template corresponding to the sound wave signal is selected from a preset feature library;

[0096] The correction error is determined based on the first matching signal in the matching template and the acoustic signal.

[0097] The acoustic signal is reconstructed based on the correction error to obtain the first reconstructed signal.

[0098] In this embodiment, when reconstructing the signal using the first reconstruction strategy, the signal quality of the acoustic signal is currently very good, possibly with only slight interference, and no adjustment is needed. However, to more accurately determine the propagation event, feature parameters can be extracted from the acoustic signal to obtain its characteristic parameters. These feature parameters may include the total duration of the entire wave, the center frequency, the rise time of the main peak, and the fall time of the main peak. To better match the template in the preset feature set, the main peak asymmetry also needs to be determined based on the rise time and fall time of the main peak. The main peak asymmetry reflects the changes in the flow field, thus allowing for more accurate determination. The system first matches a template corresponding to the actual working condition from a preset feature library. Then, based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, it selects a matching template corresponding to the acoustic signal from the preset feature library. This is to find a matching template similar to the acoustic signal from the preset feature library, select an optimization target for generating the reconstructed signal, and then use the envelope and phase characteristics of the reconstructed signal as a reference to perform linear or nonlinear calibration on the main peak amplitude, zero crossing point, etc. of the acoustic signal. That is, the correction error of each characteristic parameter is determined by the first matching signal and the acoustic signal. Then, the acoustic signal is reconstructed by the first matching signal and the correction error to obtain the first reconstructed signal.

[0099] By adopting the above scheme, the computational load is small, the response is fast, and the first reconstructed signal can be obtained quickly.

[0100] In other embodiments, to enable rapid response and determination of the matching template, before selecting a matching template corresponding to the acoustic signal from a preset feature library based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, the method further includes:

[0101] Under preset operating conditions, acquire multiple upstream and downstream acoustic signals from multiple flow points;

[0102] The upstream and downstream acoustic signals are clustered using a pre-defined clustering algorithm to obtain the reference signal for each flow point;

[0103] Feature parameters are extracted from the reference signal to obtain the reference global features, reference main peak local features, reference time series features, and reference frequency domain features, as well as the corresponding propagation time values.

[0104] A pre-defined feature library is constructed based on reference global features, reference main peak local features, reference time series features, and reference frequency domain features and their corresponding propagation time values.

[0105] In this embodiment, a corresponding preset feature library needs to be constructed to reconstruct the signal using the first reconstruction strategy. Specifically, multiple upstream and downstream acoustic signals of multiple flow points need to be acquired under preset operating conditions. The preset operating conditions can be under known and stable gas composition, temperature and pressure, and under high-precision standard wind tunnel or calibration device. The number of flow points needs to be selected to cover at least 5 flow points covering the working range of the gas acoustic flow meter, including at least zero flow, maximum flow and 3 flow points evenly distributed in the middle.

[0106] At each flow point, multiple sets of upstream and downstream acoustic signals are collected, up to 50 sets. Then, a preset clustering algorithm is used to select the 10 sets of waveforms with the most concentrated and stable morphology, and a weighted average is performed to obtain the reference signal at that flow point. As an example, the preset clustering algorithm can be K-means, which is not limited here.

[0107] After obtaining the reference signal, it is necessary to extract the characteristic parameters of the reference signal in order to construct a preset feature library. The characteristic parameters extracted from the reference signal can include reference global features, reference main peak local features, reference time series features, and reference frequency domain features and corresponding propagation time values. The reference global features can include the total duration of the whole wave and the whole wave envelope. The reference main peak local features can include the first wave amplitude, the second wave amplitude, the maximum amplitude, the main peak width, the main peak fall time and the main peak rise time. The reference time series features can include the zero-crossing time and the inter-peak duration. The reference frequency domain features can include the neutral frequency and the bandwidth.

[0108] Then, the characteristic parameters of the reference signals of all traffic points and their corresponding propagation time values ​​are stored to construct a preset feature library.

[0109] In some embodiments, a matching template corresponding to the acoustic signal is selected from a preset feature library based on the main peak asymmetry, total duration of the entire wave, and intermediate frequency, including:

[0110] The dynamic weight of each characteristic parameter is determined based on the signal quality index and preset adjustment conditions.

[0111] A weighted feature vector is constructed based on each feature parameter and its corresponding dynamic weight;

[0112] Using a preset similarity algorithm, the similarity between the weighted feature vector and the reference signal vector in the preset feature library is calculated;

[0113] Based on similarity, a matching template corresponding to the sound wave signal is selected.

[0114] In this embodiment, to reduce the possibility of matching errors due to distortion of a single feature, the dynamic weight of each feature parameter can be determined based on the signal quality index of the acoustic signal and preset adjustment conditions. For example, when the signal-to-noise ratio is low, the weight of noise-insensitive feature parameters, such as the center frequency in the frequency domain, is increased, while the weight of amplitude-type feature parameters is decreased, thus achieving dynamic weight adjustment. Then, a weighted feature vector is constructed based on the matching template and the acoustic signal, and similarity is calculated. In the process of calculating similarity, a preset similarity algorithm can be used. The preset similarity algorithm can be Mahalanobis distance or weighted cosine similarity. Mahalanobis distance takes into account the correlation between features and can calculate similarity more scientifically.

[0115] In this embodiment, the template with the smallest distance or the highest similarity can be selected as the matching template.

[0116] After obtaining the matching template, the Hilbert envelopes of the acoustic signal and the matching template can be extracted. The least squares method is used to scale the amplitude of the acoustic signal envelope and perform a small time shift to minimize the sum of squared errors between the acoustic signal envelope and the matching template envelope, thus achieving envelope alignment.

[0117] Based on envelope alignment, focus on the zero-crossing region of the main wave, calculate the phase difference between the acoustic signal and the matching template near the zero-crossing point, correct the propagation time error caused by frequency drift or phase distortion, and finally obtain the propagation time by weighted summation of the time offset of envelope alignment and the offset of phase calibration.

[0118] In other embodiments, when the signal quality index is greater than a second preset threshold and less than or equal to a first preset threshold, i.e., the reconstruction strategy adopted is the second reconstruction strategy, it indicates that the acoustic signal has high signal noise and complex distortion, and a model waveform reconstruction strategy can be used to achieve fast reconstruction. Specifically, reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal may further include:

[0119] When the reconstruction strategy is determined to be the second reconstruction strategy, the acoustic signal and the matching template are aligned at the end to obtain the serial signal;

[0120] The series signal is input into the preset reconstruction model;

[0121] The encoder of the pre-defined reconstruction model is used to capture the contextual information of the serial signal;

[0122] By using a decoder with a pre-defined reconstruction model, local details of the serial signal can be captured.

[0123] The second reconstructed signal is fitted based on contextual information and local details.

[0124] In this embodiment, before adopting the second reconstruction strategy, the preset reconstruction model needs to be trained in order to enable the preset reconstruction model to quickly and accurately reconstruct the signal.

[0125] Specifically, when training the preset reconstruction model, a large number of clean standard signals corresponding to the sound wave signal machine with various noises and distortions added can be used as the training set to train a convolutional neural network with a U-NET structure. Then, the sound wave signal collected in real time is input into the preset reconstruction model, and the preset reconstruction model outputs a purified sound wave signal, that is, the second reconstruction signal.

[0126] When inputting the acoustic signal into the preset reconstruction model, the matching template mentioned above can be used to align the beginning and end to form a serial signal of length 2N. Then, this serial signal is input into the U-NET model. The encoder in the preset reconstruction model gradually extracts the deep features of the serial signal and compresses the spatial dimension to capture the contextual information of the serial signal. The encoder can be a convolutional layer and a pooling layer in the preset reconstruction model.

[0127] The spatial dimension of the cascaded signal is then gradually recovered by the decoder of the preset reconstruction model, and the local details of the signal are accurately reconstructed by the context information captured by the encoder. The decoder of the preset reconstruction model may include deconvolution layers and upsampling layers.

[0128] After obtaining the context information and local details, signal reconstruction can be achieved to obtain the second reconstructed signal.

[0129] After obtaining the second reconstructed signal, the propagation time can be calculated directly from the second reconstructed signal.

[0130] In this embodiment, the third reconstruction strategy can be divided into two cases. One is that the signal quality index is less than or equal to the second preset threshold, but the main peak can still be detected in the acoustic signal. The other is that the signal quality index is less than or equal to the second preset threshold, but the main peak of the acoustic signal is completely submerged by noise and cannot be detected. In the case where the signal quality index is less than or equal to the second preset threshold, but the main peak can still be detected in the acoustic signal, the acoustic signal is reconstructed according to the reconstruction strategy to obtain the reconstructed signal. This may further include:

[0131] Amplification is performed on the sound wave signal to obtain an enhanced sound wave signal;

[0132] The enhanced acoustic signal is subjected to a fast Fourier transform to obtain the signal spectrum;

[0133] Determine the cross power spectrum based on the signal spectrum;

[0134] By performing weighted summation and inverse Fourier transform on the cross-power spectrum, the generalized cross-correlation function is obtained;

[0135] The reconstructed signal is determined based on the generalized cross-correlation function.

[0136] In this embodiment, the acoustic signal is first subjected to strong bandwidth filtering to retain the core frequency band. Then, the time delay of the upstream and downstream acoustic signals is calculated using the phase change weighted generalized cross-correlation method to reduce frequency-related noise interference.

[0137] Specifically, in the generalized cross-correlation method, the acoustic signal is first enhanced to obtain an enhanced acoustic signal. The filter passband can be set to ±25% of the center frequency of the ultrasonic transducer to filter out out-of-band noise and obtain the enhanced acoustic signal.

[0138] Then, a Fast Fourier Transform is performed on the enhanced acoustic signal to obtain the signal spectra Xup(f) and Xdown(f), and the cross-power spectrum is calculated using the following formula:

[0139]

[0140] in, For cross power spectrum, This indicates taking the conjugate.

[0141] The cross-power spectrum is then weighted and calculated using the following formula:

[0142]

[0143] In this method, PHAT weighting is equivalent to retaining only the aroma information of the cross power spectrum and discarding the amplitude information, so that the final cross-correlation function depends only on the phase difference between the signals and is not sensitive to amplitude changes caused by amplitude attenuation and multipath effects.

[0144] The weighted cross-power spectrum can then be calculated using the following formula:

[0145]

[0146] By performing an inverse Fourier transform on the weighted cross-power spectrum, the generalized cross-correlation function of appetite is obtained, which describes the degree of similarity between two models under different time delays.

[0147] Then, the generalized cross-correlation function is used to define a search window, and the global maximum value is found through the search window. The time delay corresponding to the global maximum value is the propagation time.

[0148] In other embodiments, if the signal quality index is less than or equal to a second preset threshold, but the main peak of the acoustic signal is completely submerged by noise and cannot be detected, the current time difference measurement can be abandoned and historical data prediction can be enabled.

[0149] Specifically, a Kalman filter or a sliding window mean predictor can be used to predict the propagation time based on the flow change trend of the previous few successful measurement cycles.

[0150] In other embodiments, when the second or third reconstruction strategy is repeatedly used under a certain working condition, these acoustic signals and the finally determined propagation time can be stored. This data can be used to fine-tune the preset reconstruction model, enabling the preset reconstruction model to continuously learn and adapt to specific complex field environments, thereby improving the accuracy of the reconstructed signals.

[0151] Based on the gas ultrasonic flow meter signal processing method provided in the above embodiments, this application also provides specific implementation methods of the gas ultrasonic flow meter signal processing device. Please refer to the following embodiments.

[0152] First see Figure 2 The gas ultrasonic flow meter signal processing device 200 provided in this application embodiment may include:

[0153] Acquisition module 201 is used to acquire the acoustic signal of the target flow point;

[0154] The determination module 202 is used to determine the signal-to-noise ratio, main peak salience, and waveform coherence of the acoustic signal based on the acoustic signal. The signal-to-noise ratio is used to evaluate the degree of noise contamination of the signal, the main peak salience is used to evaluate the clarity of the main peak, and the waveform coherence is used to evaluate the symmetry of the flow field.

[0155] The determination module 202 is also used to determine the signal quality index of the acoustic signal based on the signal-to-noise ratio, the salience of the main peak, and the waveform coherence;

[0156] The determination module 202 is also used to determine the reconstruction strategy of the acoustic signal based on the signal quality index and the preset decision range;

[0157] Signal reconstruction module 203 is used to reconstruct the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal;

[0158] The determination module 202 is also used to determine the propagation time based on the reconstructed signal in order to determine the gas flow rate at the target flow point.

[0159] As an alternative implementation, the acquisition module 201 can also be used for:

[0160] The characteristic parameters of each gas ultrasonic flow meter are obtained, including resonant frequency, dynamic resistance and mechanical quality factor;

[0161] The overall matching index is determined based on the resonant frequency, dynamic resistance, and mechanical quality factor, as well as their corresponding weight values.

[0162] Match the gas ultrasonic flow meters whose comprehensive matching index is lower than the preset threshold to obtain a gas ultrasonic flow meter combination.

[0163] Module 201 is specifically used for:

[0164] Acquire the acoustic signal of the gas ultrasonic flow meter combination at the target flow point.

[0165] As an optional implementation, the reconstruction strategy includes a first reconstruction strategy, which includes a parameterized fine-tuning reconstruction strategy. The reconstruction signal includes a first reconstruction signal. The signal reconstruction module 203 can also be used for:

[0166] When the reconstruction strategy is determined to be the first reconstruction strategy, feature parameters are extracted based on the acoustic signal. The feature parameters include the total duration of the whole wave, the center frequency, the rise time of the main peak, and the fall time of the main peak.

[0167] The asymmetry of the main peak is determined based on the rise time and fall time of the main peak. The asymmetry of the main peak is used to reflect the changes in the flow field.

[0168] Based on the asymmetry of the main peak, the total duration of the whole wave, and the intermediate frequency, a matching template corresponding to the sound wave signal is selected from a preset feature library;

[0169] The correction error is determined based on the first matching signal in the matching template and the acoustic signal.

[0170] The acoustic signal is reconstructed based on the correction error to obtain the first reconstructed signal.

[0171] As an alternative implementation, the signal reconstruction module 203 can also be used for:

[0172] Under preset operating conditions, acquire multiple upstream and downstream acoustic signals from multiple flow points;

[0173] The upstream and downstream acoustic signals are clustered using a pre-defined clustering algorithm to obtain the reference signal for each flow point;

[0174] Feature parameters are extracted from the reference signal to obtain the reference global features, reference main peak local features, reference time series features, and reference frequency domain features, as well as the corresponding propagation time values.

[0175] A pre-defined feature library is constructed based on reference global features, reference main peak local features, reference time series features, and reference frequency domain features and their corresponding propagation time values.

[0176] As an alternative implementation, the signal reconstruction module 203 can also be used for:

[0177] The dynamic weight of each characteristic parameter is determined based on the signal quality index and preset adjustment conditions.

[0178] A weighted feature vector is constructed based on each feature parameter and its corresponding dynamic weight;

[0179] Using a preset similarity algorithm, the similarity between the weighted feature vector and the reference signal vector in the preset feature library is calculated;

[0180] Based on similarity, a matching template corresponding to the sound wave signal is selected.

[0181] As an optional implementation, the reconstruction strategy includes a second reconstruction strategy, which includes a model waveform reconstruction strategy. The reconstructed signal includes a second reconstructed signal. The signal reconstruction module 203 can also be used for:

[0182] When the reconstruction strategy is determined to be the second reconstruction strategy, the acoustic signal and the matching template are aligned at the end to obtain the serial signal;

[0183] The series signal is input into the preset reconstruction model;

[0184] The encoder of the pre-defined reconstruction model is used to capture the contextual information of the serial signal;

[0185] By using a decoder with a pre-defined reconstruction model, local details of the serial signal can be captured.

[0186] The second reconstructed signal is fitted based on contextual information and local details.

[0187] As an optional implementation, the reconstruction strategy includes a third reconstruction strategy, which includes a conservative correlation reconstruction strategy. The reconstruction signal includes a third reconstruction signal. The signal reconstruction module 203 can also be used for:

[0188] Amplification is performed on the sound wave signal to obtain an enhanced sound wave signal;

[0189] The enhanced acoustic signal is subjected to a fast Fourier transform to obtain the signal spectrum;

[0190] Determine the cross power spectrum based on the signal spectrum;

[0191] By performing weighted summation and inverse Fourier transform on the cross-power spectrum, the generalized cross-correlation function is obtained;

[0192] The reconstructed signal is determined based on the generalized cross-correlation function.

[0193] Figure 3 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0194] An electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0195] Specifically, the processor 301 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0196] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 302 may include removable or non-removable (or fixed) media, or memory 302 may be non-volatile solid-state memory. Memory 302 may be internal or external to the integrated gateway disaster recovery device.

[0197] In one instance, memory 302 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0198] Memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the gas ultrasonic flowmeter signal processing method according to the first aspect of this disclosure.

[0199] The processor 301 reads and executes computer program instructions stored in the memory 302 to achieve... Figure 1 A signal processing method for a gas ultrasonic flow meter in the embodiment shown.

[0200] In one example, the electronic device may also include a communication interface 303 and a bus 304. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 304 and complete communication with each other.

[0201] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0202] Bus 304 includes hardware, software, or both, that couples components of an electronic device together. For example, and not as a limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 304 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0203] This electronic device can execute the gas ultrasonic flow meter signal processing method in the embodiments of this application, thereby achieving the combination Figures 1-2 The method and apparatus for signal processing of a gas ultrasonic flow meter are described.

[0204] Furthermore, in conjunction with the gas ultrasonic flow meter signal processing method in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the gas ultrasonic flow meter signal processing methods in the above embodiments.

[0205] In an optional embodiment, in conjunction with the gas ultrasonic flow meter signal processing method in the above embodiments, this application embodiment can provide a computer program product to implement it. The instructions in the computer program product are executed by the processor of an electronic device, enabling the electronic device to implement any of the gas ultrasonic flow meter signal processing methods in the above embodiments.

[0206] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0207] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0208] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0209] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0210] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A signal processing method for a gas ultrasonic flow meter, characterized in that, include: Acquire the acoustic signal at the target flow point; Based on the acoustic signal, the signal-to-noise ratio, peak salience, and waveform coherence of the acoustic signal are determined. The signal-to-noise ratio is used to assess the degree to which the signal is contaminated by noise, the peak salience is used to assess the clarity of the peak, and the waveform coherence is used to assess the symmetry of the flow field. The signal quality index of the acoustic signal is determined based on the signal-to-noise ratio, the salience of the main peak, and the waveform coherence. Based on the signal quality index and the preset decision range, a reconstruction strategy for the acoustic signal is determined; The acoustic signal is reconstructed according to the reconstruction strategy to obtain a reconstructed signal; Based on the reconstructed signal, the propagation time is determined to determine the gas flow rate at the target flow point; The reconstruction strategy includes a first reconstruction strategy, which includes a parameterized fine-tuning reconstruction strategy. The reconstruction signal includes a first reconstruction signal. The step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes: When the reconstruction strategy is determined to be the first reconstruction strategy, feature parameters are extracted based on the acoustic signal. The feature parameters include the total duration of the whole wave, the center frequency, the rise time of the main peak, and the fall time of the main peak. The asymmetry of the main peak is determined based on the rise time and fall time of the main peak, and the asymmetry of the main peak is used to reflect the changes in the flow field. Based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, a matching template corresponding to the acoustic signal is selected from a preset feature library; The correction error is determined based on the first matching signal in the matching template and the acoustic signal; The acoustic signal is reconstructed based on the correction error to obtain a first reconstructed signal.

2. The method according to claim 1, characterized in that, Before acquiring the acoustic signal of the target flow point, the method further includes: The characteristic parameters of each gas ultrasonic flow meter are obtained, including resonant frequency, dynamic resistance and mechanical quality factor; Based on the resonant frequency, dynamic resistance, and mechanical quality factor, as well as the corresponding weight values, the comprehensive matching index is determined. The gas ultrasonic flow meters whose comprehensive matching index is lower than the preset threshold are matched to obtain a gas ultrasonic flow meter combination. The acquisition of the acoustic signal at the target flow point includes: Acquire the acoustic signal of the gas ultrasonic flow meter assembly at the target flow point.

3. The method according to claim 1, characterized in that, Before selecting a matching template corresponding to the acoustic signal from a preset feature library based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, the method further includes: Under preset operating conditions, acquire multiple upstream and downstream acoustic signals from multiple flow points; The upstream and downstream acoustic signals are clustered using a preset clustering algorithm to obtain the reference signal for each flow point; Feature parameters are extracted from the reference signal to obtain reference global features, reference main peak local features, reference time series features, and reference frequency domain features, as well as the corresponding propagation time values. Based on the aforementioned global reference features, local reference main peak features, time-series reference features, and frequency-domain reference features, along with their corresponding propagation time values, a preset feature library is constructed.

4. The method according to claim 3, characterized in that, The step of selecting a matching template corresponding to the acoustic signal from a preset feature library based on the main peak asymmetry, total duration of the full wave, and intermediate frequency includes: Based on the signal quality index and preset adjustment conditions, determine the dynamic weight of each feature parameter; A weighted feature vector is constructed based on each feature parameter and its corresponding dynamic weight. The similarity between the weighted feature vector and the reference signal vector in the preset feature library is calculated using a preset similarity algorithm. Based on the similarity, a matching template corresponding to the sound wave signal is selected.

5. The method according to claim 1, characterized in that, The reconstruction strategy includes a second reconstruction strategy, which includes a model waveform reconstruction strategy. The reconstructed signal includes a second reconstructed signal. The step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes: When the reconstruction strategy is determined to be the second reconstruction strategy, the acoustic signal and the matching template are aligned at the end to obtain a serial signal; The series signal is input into a preset reconstruction model; The encoder of the preset reconstruction model is used to capture the context information of the serial signal; The decoder of the preset reconstruction model is used to capture local details of the serial signal; The second reconstructed signal is fitted based on the context information and the local details.

6. The method according to claim 1, characterized in that, The reconstruction strategy includes a third reconstruction strategy, which includes a conservative correlation reconstruction strategy. The reconstructed signal includes a third reconstructed signal. The step of reconstructing the acoustic signal according to the reconstruction strategy to obtain the reconstructed signal includes: The acoustic signal is enhanced to obtain an enhanced acoustic signal; Perform a Fast Fourier Transform on the enhanced acoustic signal to obtain the signal spectrum; Based on the signal spectrum, determine the cross-power spectrum; The cross-power spectrum is weighted and subjected to inverse Fourier transform to obtain the generalized cross-correlation function; The reconstructed signal is determined based on the generalized cross-correlation function.

7. A signal processing device for a gas ultrasonic flow meter, characterized in that, The device includes: The acquisition module is used to acquire the acoustic signal of the target flow point; The determination module is used to determine the signal-to-noise ratio, peak salience, and waveform coherence of the acoustic signal based on the acoustic signal. The signal-to-noise ratio is used to assess the degree to which the signal is contaminated by noise, the peak salience is used to assess the clarity of the peak, and the waveform coherence is used to assess the symmetry of the flow field. The determining module is also used to determine the signal quality index of the acoustic signal based on the signal-to-noise ratio, main peak salience, and waveform coherence. The determining module is also used to determine a reconstruction strategy for the acoustic signal based on the signal quality index and a preset decision range; A signal reconstruction module is used to reconstruct the acoustic signal according to the reconstruction strategy to obtain a reconstructed signal; The determining module is further configured to determine the propagation time based on the reconstructed signal, so as to determine the gas flow rate at the target flow point; The reconstruction strategy includes a first reconstruction strategy, the first reconstruction strategy includes a parameterized fine-tuning reconstruction strategy, the reconstruction signal includes a first reconstruction signal, and the signal reconstruction module is further specifically used for: When the reconstruction strategy is determined to be the first reconstruction strategy, feature parameters are extracted based on the acoustic signal. The feature parameters include the total duration of the whole wave, the center frequency, the rise time of the main peak, and the fall time of the main peak. The asymmetry of the main peak is determined based on the rise time and fall time of the main peak, and the asymmetry of the main peak is used to reflect the changes in the flow field. Based on the main peak asymmetry, total duration of the full wave, and intermediate frequency, a matching template corresponding to the acoustic signal is selected from a preset feature library; The correction error is determined based on the first matching signal in the matching template and the acoustic signal; The acoustic signal is reconstructed based on the correction error to obtain a first reconstructed signal.

8. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the gas ultrasonic flow meter signal processing method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the gas ultrasonic flow meter signal processing method as described in any one of claims 1-6.

Images