Intelligent acquisition and processing method for data of gate flowmeter

Abstract

This invention relates to the field of data measurement technology, specifically to an intelligent data acquisition and processing method for gate flowmeters. The method involves: obtaining the main period duration based on the spectral distribution characteristics of the component signals of the flow velocity time series; obtaining sliding windows at different positions in the component signals based on the main period duration; obtaining continuous data characteristic values ​​based on the differences in data change trends between adjacent sliding windows; obtaining overall stable characteristic values ​​based on the differences in data change trends between different sliding windows; obtaining the component comprehensive reliability based on the continuous data characteristic values ​​and the overall stable characteristic values; obtaining a reliability coefficient based on the differences in component comprehensive reliability and the differences in the main period duration; and obtaining the energy intensity based on the amplitude characteristics of the component signals. This invention obtains the importance level based on the reliability coefficient and energy intensity; and obtains the fused flow velocity time series based on the component signals and their corresponding importance levels, thereby improving the accuracy of flow measurement.

Description

Technical Field

[0001] This invention relates to the field of data measurement technology, and specifically to a method for intelligent data acquisition and processing of gate flowmeters. Background Technology

[0002] Gate flow meters are widely used in agricultural irrigation, urban water supply and drainage, and water resource allocation. Utilizing ultrasonic principles for flow measurement is one of the mainstream technologies in the field of culvert gates. Typically, multiple sets of ultrasonic transducers are deployed within the gate culvert to create multiple flow measurement channels, measuring the flow velocity within the cross-section and calculating the actual flow rate of the gate based on water level information. As the measurement structure of gate flow meters evolves from single-channel to multi-channel and three-dimensional, it can acquire richer measurement data under complex flow conditions. However, this also significantly increases the complexity of data acquisition and processing.

[0003] When using a gate flowmeter with a multi-channel ultrasonic flow measurement structure to measure flow, although it can acquire multiple flow velocity data from different spatial locations, the measurement stability and reliability vary significantly under different operating conditions because different channels are susceptible to factors such as local flow inhomogeneity, turbulence, bubbles, and gate opening and closing disturbances. If low-reliability channel data is not effectively suppressed and there is a lack of a real-time evaluation and differentiation mechanism for the reliability of individual channel measurement data, and only existing fixed weight or simple averaging fusion methods are used, the fusion results are easily interfered with by abnormal data, thereby affecting the accuracy and stability of flow calculation. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention aims to provide an intelligent data acquisition and processing method for gate flow meters, the specific technical solution of which is as follows:

[0005] Obtain the flow velocity time sequence measured by ultrasonic transducers at different locations in the flow cross section;

[0006] The flow velocity time series is decomposed using the EEMD algorithm to obtain different component signals; the main period duration is obtained based on the spectral distribution characteristics of the component signals; sliding windows at different positions in the component signals are obtained based on the main period duration; continuous data feature values ​​are obtained based on the differences in data change trends between adjacent sliding windows; overall stable feature values ​​are obtained based on the differences in data change trends between different sliding windows; and the overall component reliability is obtained based on the continuous data feature values ​​and the overall stable feature values.

[0007] The reliability coefficient of the component signal is obtained based on the differences in the overall reliability of the different component signals and the differences in the duration of the main period; the energy intensity is obtained based on the amplitude characteristics of the component signal; and the importance is obtained based on the reliability coefficient and energy intensity of the component signal.

[0008] The fusion flow rate timing is obtained based on the component signals and their corresponding importance.

[0009] Further, the step of obtaining the duration of the main period based on the spectral distribution characteristics of the component signal includes:

[0010] The spectral data of the component signal is obtained by short-time Fourier transform, where T represents the duration of the main period of the component signal. This represents the maximum frequency in the spectrum data. This represents the minimum frequency in the spectral data, and s represents different frequencies. This represents the spectral amplitude corresponding to the s-th frequency.

[0011] Further, the step of obtaining the sliding window at different positions in the component signal based on the main period duration includes:

[0012] A timing window with a length equal to the duration of the main cycle is constructed and slid from the starting position of the component signal with a preset step size, and a sliding window is obtained for each slide.

[0013] Furthermore, the step of obtaining continuous feature values ​​of the data based on the differences in the data change trends of adjacent sliding windows includes:

[0014] The average dynamic time-warped distance between all adjacent sliding windows in the component signal is calculated and negatively correlated to obtain the continuous feature values ​​of the data.

[0015] Furthermore, the step of obtaining the overall stable feature value based on the differences in data change trends across different sliding windows includes:

[0016] The average value of the dynamic time warping distance between any two sliding windows in the component signal is calculated and negatively correlated to obtain the overall stable characteristic value.

[0017] Further, the step of obtaining the component comprehensive reliability based on the continuous characteristic values ​​of the data and the overall stable characteristic values ​​includes:

[0018] The component comprehensive reliability of the component signal is obtained by calculating the product of the continuous feature value of the data and the overall stable feature value.

[0019] Furthermore, the step of obtaining the reliability coefficient of the component signal based on the difference characteristics of the component comprehensive reliability and the difference characteristics of the main period duration of different component signals includes:

[0020] In the formula Let represent the confidence coefficient of the 'a'-th component signal, and g represent the number of component signals in all flow velocity time series. This represents the overall reliability of the a-th component signal. This represents the overall reliability of the b-th other component signal. This represents the duration of the principal period of the a-th component signal. This represents the duration of the principal period of the b-th component signal. This represents an exponential function with the natural constant as its base.

[0021] Further, the step of obtaining the energy intensity based on the amplitude characteristics of the component signal includes:

[0022] In the formula, E represents the energy intensity of the component signal, and N represents the number of sampling points of the component signal. This represents the amplitude corresponding to the nth sampling point.

[0023] Furthermore, the step of obtaining the importance based on the confidence coefficient and energy intensity of the component signal includes:

[0024] The importance of the component signal is obtained by multiplying the confidence coefficient of the component signal by the energy intensity.

[0025] Further, the step of obtaining the fused flow rate timing based on the component signals and their corresponding importance includes:

[0026] In the formula This represents the fused flow rate timing sequence, where M represents the number of flow rate timing sequences and H represents the number of component signals in the flow rate timing sequence. This represents the h-th component signal in the flow velocity time series. This indicates the importance of the h-th component signal. This represents the reconstructed flow velocity time series of the m-th flow velocity time series.

[0027] The present invention has the following beneficial effects:

[0028] In this invention, acquiring component signals enables the analysis of disturbance characteristics in the velocity time series; acquiring the duration of the main period enables the analysis of the main change period of the component signals, making the selection of the subsequent sliding window more accurate and improving the accuracy of component signal stability analysis. The sliding window can be used to analyze the stability characteristics of the component signals in the time dimension, thereby determining the reliability of the velocity characteristics represented by the component signals. Acquiring continuous feature values ​​of the data can characterize the local change stability characteristics of the component signals; acquiring overall stable feature values ​​can characterize the overall change stability characteristics of the component signals; acquiring the component comprehensive reliability can characterize the reliability of the component signals. Acquiring the confidence coefficient can characterize the relative reliability of the component signals based on the differences in the component comprehensive reliability of the component signals of all velocity time series, thereby determining the contribution degree of different component signals during fusion. Acquiring the energy intensity can characterize the importance of the component signals in representing the flow characteristics. Acquiring the importance degree can more accurately determine the weight of different component signals during fusion based on the confidence coefficient and energy intensity. Finally, obtaining the fused velocity time series based on the component signals and their corresponding importance degrees can reduce the impact of local disturbances on the accuracy of flow measurement during the measurement process, improving the accuracy of flow measurement. Attached Figure Description

[0029] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 The flowchart illustrates a method for intelligent data acquisition and processing of a gate flow meter, as provided in one embodiment of the present invention. Detailed Implementation

[0031] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a gate flowmeter data intelligent acquisition and processing method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent data acquisition and processing method for gate flowmeters provided by this invention.

[0034] Please see Figure 1 The diagram illustrates a flowchart of an intelligent data acquisition and processing method for a gate flow meter according to an embodiment of the present invention. The method includes the following steps:

[0035] Step S1: Obtain the flow velocity timing measured by ultrasonic transducers at different locations in the flow cross section.

[0036] In this embodiment of the invention, the implementation scenario involves measuring flow velocity using multiple ultrasonic transducers to improve the accuracy of flow velocity and flow rate measurements. First, the flow velocity time sequence measured by ultrasonic transducers at different locations within the flow cross-section is obtained. The placement and number of ultrasonic transducers can be determined by the implementer based on the implementation scenario. The ultrasonic transducers in each channel must maintain synchronous acquisition to obtain the flow velocity time sequence measured at different locations within the same time period.

[0037] Step S2: Decompose the flow velocity time series using the EEMD algorithm to obtain different component signals; obtain the main period duration based on the spectral distribution characteristics of the component signals; obtain sliding windows at different positions in the component signals based on the main period duration; obtain continuous data feature values ​​based on the differences in data change trends between adjacent sliding windows; obtain overall stable feature values ​​based on the differences in data change trends between different sliding windows; and obtain the component comprehensive reliability based on the continuous data feature values ​​and the overall stable feature values.

[0038] When using a gate flowmeter with a multi-channel ultrasonic flow measurement structure for flow measurement, although it can acquire multiple velocity measurement data from different locations in space, the complex flow regime within the gate's cross-section means that different channels are easily affected by factors such as uneven local velocity distribution, turbulence, bubble inclusion, and transient disturbances caused by gate opening and closing during actual operation. This leads to significant differences in the stability and reliability of the measurement data from each channel at different times. Furthermore, even for the same channel, the degree of disturbance to the acquired ultrasonic flow measurement signal varies at different time scales or frequency components. For example, low-frequency components usually reflect the overall flow state of the gate and slow velocity changes caused by water level changes, exhibiting relatively high stability; mid-frequency components are mostly related to local flow regime adjustments or changes in gate operating conditions; while high-frequency components are often related to disturbances such as bubbles, local turbulence, and random noise, resulting in relatively poor stability. If the velocity time sequence of each channel is directly fused with fixed weights, local abnormal disturbance components can easily have a disproportionate impact on the final flow calculation result, thereby reducing the accuracy and stability of the measurement results. Therefore, it is necessary to decompose the ultrasonic flow measurement signal of each channel into several component signals with different variation characteristics, and analyze each component signal separately. Through adaptive fusion and reconstruction of multi-channel and multi-component flow measurement signals, the real flow characteristics within the flow cross section can be extracted more accurately, thereby improving the accuracy and reliability of flow measurement.

[0039] Furthermore, the flow velocity time series is decomposed using the EEMD algorithm to obtain different component signals. EEMD is an improved version of Empirical Mode Decomposition (EMD), which can improve the stability and physical meaning of the decomposition. This algorithm is existing technology, and the specific steps will not be elaborated here. Since the different component signals obtained after the flow velocity time series of each channel is decomposed represent different frequency components, and the variation characteristics of different frequency component signals are also different during analysis, it is necessary to obtain the main changing frequency and main period duration through the corresponding spectral information in order to obtain a suitable analysis scale for each component signal. Therefore, the main period duration is obtained based on the spectral distribution characteristics of the component signals. Preferably, in this embodiment of the invention, the step of obtaining the main period duration includes:

[0040]

[0041] The spectral data of the component signal is obtained through short-time Fourier transform. It should be noted that short-time Fourier transform is an existing technology, and the specific steps will not be elaborated further. In the formula, T represents the duration of the principal period of the component signal. This represents the maximum frequency in the spectrum data. This represents the minimum frequency in the spectral data, and s represents different frequencies. This represents the spectral amplitude corresponding to the s-th frequency. It represents the weighted average of the frequency of the component signal, which can reflect the main frequency of change, and thus the duration of the main period can be obtained based on the reciprocal of the main frequency of change.

[0042] After obtaining the main period duration of the component signal, sliding windows at different positions in the component signal can be obtained based on the main period duration. Preferably, in this embodiment of the invention, the step of obtaining the sliding window includes: constructing a time-series window with a length equal to the main period duration and sliding it from the starting position of the component signal with a preset step size, obtaining one sliding window each time. In this embodiment of the invention, the preset step size is half of the main period duration, which can be determined by the implementer according to the implementation scenario. The main period duration can be used to determine a suitable window size, which facilitates the analysis of the stability of the component signal. In the actual operation of the gate flowmeter, when the gate is under relatively stable operating conditions, the water flow characteristics in the flow cross-section usually exhibit continuous and gentle evolution characteristics in the time dimension. Correspondingly, each component signal should show a continuous trend of change, limited fluctuation amplitude, and relatively stable overall in the time dimension. If the variation characteristics of a component signal differ significantly between adjacent time periods, such as sudden amplitude jumps or increased high-frequency fluctuations, it usually indicates that the component is affected by random factors such as bubble doping, local turbulence disturbances, or measurement noise during that time period, resulting in poor temporal continuity and relatively low reliability in representing the true flow velocity state. Furthermore, if the component signal exhibits significant fluctuations or poor overall fluctuation uniformity over a longer time range, it means that the overall stability of the component is insufficient, and its measurement results are more susceptible to abnormal disturbances. Therefore, the evaluation values ​​of both local and overall time periods constitute important criteria for judging the reliability of component signals. First, continuous characteristic values ​​of the data are obtained based on the differences in the data variation trends of adjacent sliding windows.

[0043] Preferably, in this embodiment of the invention, the step of obtaining continuous data feature values ​​includes: calculating the average dynamic time warping distance between all adjacent sliding windows in the component signal and performing negative correlation mapping to obtain continuous data feature values. It should be noted that the dynamic time warping distance is obtained using existing dynamic time warping algorithms. The more similar the changing trends of two sequences, the smaller the dynamic time warping distance. Therefore, a larger continuous data feature value means better continuity of the component signal over time. In this embodiment of the invention, through... The function performs a negative correlation mapping. Let X represent an exponential function with the natural constant as its base, and let X represent the mapping object. Further, for the overall stability of the component signal throughout the entire acquisition period, an overall stability feature value can be obtained based on the differences in the data change trends of different sliding windows. Preferably, in this embodiment of the invention, the step of obtaining the overall stability feature value includes: calculating the average value of the dynamic time warping distance between any two sliding windows in the component signal and performing a negative correlation mapping to obtain the overall stability feature value. A larger overall stability feature value indicates better overall stability of the component signal throughout the entire acquisition period.

[0044] Furthermore, the component comprehensive reliability can be obtained based on the continuous characteristic values ​​and the overall stable characteristic values ​​of the data. Preferably, in this embodiment of the invention, the step of obtaining the component comprehensive reliability includes: calculating the product of the continuous characteristic values ​​of the data and the overall stable characteristic values ​​to obtain the component comprehensive reliability of the component signal. The higher the component comprehensive reliability, the higher the reliability of the flow velocity data measured by the component signal.

[0045] Step S3: Obtain the reliability coefficient of the component signal based on the differences in the overall reliability of the different component signals and the differences in the duration of the main period; obtain the energy intensity based on the amplitude characteristics of the component signal; and obtain the importance based on the reliability coefficient and energy intensity of the component signal.

[0046] The duration of the dominant change period of a component signal can characterize the dominant change timescale of that component signal. It is inversely related to the dominant change frequency and reflects the change pattern of the flow velocity characteristic corresponding to that component in the time dimension. When two component signals from different channels or different decomposition levels have similar durations of their dominant change periods, they usually indicate that they may reflect the same or similar water flow change characteristics, possessing high comparability and a basis for fusion. In this case, if the overall reliability index of one component signal is higher than that of the other, it indicates that the representation of the water flow state by that component signal is more reliable during the acquisition process, and it should be given higher weight in the subsequent component fusion process to enhance its contribution to the final fused and reconstructed signal. Therefore, the reliability coefficient of the component signal is obtained based on the differences in the overall reliability and the differences in the duration of the dominant period of different component signals. Preferably, in this embodiment of the invention, the step of obtaining the reliability coefficient includes:

[0047]

[0048] In the formula, Let represent the confidence coefficient of the 'a'-th component signal, and g represent the number of component signals in all flow velocity time series. This represents the overall reliability of the a-th component signal. This represents the overall reliability of the b-th other component signal. This represents the duration of the principal period of the a-th component signal. This represents the duration of the principal period of the b-th component signal. This represents an exponential function with base to the natural constant. When The larger the value, the stronger the reliability of the a-th component signal compared to the other b-th component signals. The principal period duration is used to determine the comparability of component signals. The more similar the principal period durations of two component signals, the more reliable their overall component reliability ratio is, and thus... The weights are used as a ratio. A higher confidence coefficient means that the velocity data of that component signal is relatively more reliable.

[0049] Furthermore, the energy intensity of a component signal reflects its amplitude and variation contribution to the overall flow measurement signal. When the component energy is large and its reliability is high, it usually means that the component signal has greater importance in characterizing water flow features and should be retained and enhanced as a major contributing component signal during the weighted fusion reconstruction process. Therefore, the energy intensity is obtained based on the amplitude characteristics of the component signal. Preferably, in this embodiment of the invention, the step of obtaining the energy intensity includes:

[0050]

[0051] In the formula, E represents the energy intensity of the component signal, and N represents the number of sampling points of the component signal. This represents the amplitude corresponding to the nth sampling point. A higher energy intensity indicates a more important signal component, and therefore, it should be preserved during component reconstruction.

[0052] After obtaining the energy intensity and confidence coefficient of each component signal, the importance can be determined based on the confidence coefficient and energy intensity of the component signal. Preferably, in this embodiment of the invention, the step of obtaining the importance includes: calculating the product of the confidence coefficient and the energy intensity of the component signal to obtain the importance of the component signal. The greater the importance, the higher the importance of the component signal during reconstruction, and the more the water flow characteristics reflected by the component signal need to be preserved.

[0053] Step S4: Obtain the fusion flow rate timing based on the component signals and their corresponding importance.

[0054] After obtaining the importance of each component signal, the fused flow rate timing can be obtained based on the component signals and their corresponding importance. Preferably, in this embodiment of the invention, the step of obtaining the fused flow rate timing includes:

[0055]

[0056] In the formula, This represents the fused flow rate timing sequence, where M represents the number of flow rate timing sequences and H represents the number of component signals in the flow rate timing sequence. This represents the h-th component signal in the flow velocity time series. This indicates the importance of the h-th component signal. This represents the reconstructed velocity time series for the m-th velocity sequence. It should be noted that component signal reconstruction is a current technology. The greater the importance of the component signal, the more its data is preserved during reconstruction. Reconstruction improves the accuracy and reliability of the velocity time series in characterizing water flow features. Finally, the mean of all reconstructed velocity time series is calculated, accurately representing the overall velocity variation characteristics of the cross-section, improving measurement accuracy, and effectively suppressing the influence of local abnormal disturbances on the measurement results. Subsequent statistical analysis of velocity and flow rate can be performed based on this reconstructed velocity time series.

[0057] In summary, this invention provides an intelligent data acquisition and processing method for gate flowmeters. It obtains the main period duration based on the spectral distribution characteristics of the component signals of the flow velocity time series; obtains sliding windows at different positions in the component signals based on the main period duration; obtains continuous data feature values ​​based on the differences in data change trends between adjacent sliding windows; obtains overall stable feature values ​​based on the differences in data change trends between different sliding windows; obtains component comprehensive reliability based on the continuous data feature values ​​and overall stable feature values; obtains a reliability coefficient based on the differences in component comprehensive reliability and the differences in main period duration; and obtains energy intensity based on the amplitude characteristics of the component signals. This invention obtains importance based on the reliability coefficient and energy intensity; and obtains fused flow velocity time series based on the component signals and their corresponding importance, thus improving the accuracy of flow measurement.

[0058] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for intelligent data acquisition and processing of gate flowmeters, characterized in that, The method includes the following steps: Obtain the flow velocity time sequence measured by ultrasonic transducers at different locations in the flow cross section; The flow velocity time series is decomposed using the EEMD algorithm to obtain different component signals; the main period duration is obtained based on the spectral distribution characteristics of the component signals; sliding windows at different positions in the component signals are obtained based on the main period duration; continuous data feature values ​​are obtained based on the differences in data change trends between adjacent sliding windows; overall stable feature values ​​are obtained based on the differences in data change trends between different sliding windows; and the overall component reliability is obtained based on the continuous data feature values ​​and the overall stable feature values. The reliability coefficient of the component signal is obtained based on the differences in the overall reliability of the different component signals and the differences in the duration of the main period; the energy intensity is obtained based on the amplitude characteristics of the component signal; and the importance is obtained based on the reliability coefficient and energy intensity of the component signal. The fusion flow rate timing is obtained based on the component signals and their corresponding importance.

2. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the duration of the main period based on the spectral distribution characteristics of the component signal includes: The spectral data of the component signal is obtained by short-time Fourier transform, where T represents the duration of the main period of the component signal. This represents the maximum frequency in the spectrum data. This represents the minimum frequency in the spectral data, and s represents different frequencies. This represents the spectral amplitude corresponding to the s-th frequency.

3. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the sliding window at different positions in the component signal based on the main period duration includes: A timing window with a length equal to the duration of the main cycle is constructed and slid from the starting position of the component signal with a preset step size, and a sliding window is obtained for each slide.

4. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining continuous feature values ​​of data based on the difference characteristics of data change trends in adjacent sliding windows includes: The average dynamic time-warped distance between all adjacent sliding windows in the component signal is calculated and negatively correlated to obtain the continuous feature values ​​of the data.

5. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the overall stable feature value based on the differences in data change trends of different sliding windows includes: The average value of the dynamic time warping distance between any two sliding windows in the component signal is calculated and negatively correlated to obtain the overall stable characteristic value.

6. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the component comprehensive reliability based on the continuous characteristic value of the data and the overall stable characteristic value includes: The component comprehensive reliability of the component signal is obtained by calculating the product of the continuous feature value of the data and the overall stable feature value.

7. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the reliability coefficient of the component signal based on the differences in the overall reliability of different component signals and the differences in the duration of the main period includes: In the formula Let represent the confidence coefficient of the 'a'-th component signal, and g represent the number of component signals in all flow velocity time series. This represents the overall reliability of the a-th component signal. This represents the overall reliability of the b-th other component signal. This represents the duration of the principal period of the a-th component signal. This represents the duration of the principal period of the b-th component signal. This represents an exponential function with the natural constant as its base.

8. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the energy intensity based on the amplitude characteristics of the component signal includes: In the formula, E represents the energy intensity of the component signal, and N represents the number of sampling points of the component signal. This represents the amplitude corresponding to the nth sampling point.

9. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the importance based on the confidence coefficient and energy intensity of the component signal includes: The importance of the component signal is obtained by multiplying the confidence coefficient of the component signal by the energy intensity.

10. The intelligent data acquisition and processing method for a gate flow meter according to claim 1, characterized in that, The step of obtaining the fused flow rate timing based on the component signals and their corresponding importance includes: In the formula This represents the fused flow rate timing sequence, where M represents the number of flow rate timing sequences and H represents the number of component signals in the flow rate timing sequence. This represents the h-th component signal in the flow velocity time series. This indicates the importance of the h-th component signal. This represents the reconstructed flow velocity time series of the m-th flow velocity time series.

Images