Channel flow measurement method of Burg algorithm based on Zoom-FFT and AR model

Abstract

The invention discloses a channel flow measurement method based on a Burg algorithm of a Zoom-FFT (Fast Fourier Transform) and an AR (Augmented Reality) model. The measuring method comprises the following steps: (1) the radar transceiver transmits frequency-modulated continuous waves; (2) receiving an echo signal by the radar flowmeter, and mixing a transmitted wave with an echo to obtain an intermediate frequency signal; (3) the water level gauge module carries out band selection Fourier transform (Zoom-FFT) on the intermediate frequency signal and carries out frequency spectrum estimation on the intermediate frequency signal; (4) the current meter module performs frequency spectrum estimation on the intermediate frequency signal by using a Burg algorithm of an AR model; and (5) respectively obtaining water level and flow velocity data after spectrum conversion according to the highest peak, and converting the flow according to a flow velocity area method model. According to the flow calculation model and the radar signal processing technology, the two important parameters of the water level and the flow velocity are obtained through the selected-band Fourier transform and the AR model Burg algorithm, the water level measurement precision and the flow velocity measurement precision are further improved, and the method has great significance in channel flow measurement and reasonable application of water resources.

Description

Technical Field

[0001] The present invention relates to the field of signal processing, and in particular to a channel flow measurement method, and in particular to a channel flow measurement method based on a Burg algorithm of a Zoom-FFT and AR model. Background Art

[0002] In the 21st century, water resources are an important resource for human production and life. With the rapid growth of social economy and population, the shortage of water resources has become increasingly prominent. As a major water user, agriculture has been facing increasing water supply and demand issues. As the main agricultural water user, irrigation areas are an important foundation for ensuring national food security and an important guarantee for the development of agriculture and rural economy. Therefore, the rational optimization of irrigation channel water resources plays an important role in improving water resource utilization, promoting the sustainable development and utilization of agricultural water resources, and balancing the relationship between irrigation channel water resources economy and risk.

[0003] There are two main methods for measuring channel flow at present: hydraulic method and velocity-area method. The hydraulic method uses the relationship between water level and flow to calculate the flow from the measured water level through hydraulic formulas, which is not affected by human factors and river characteristics. The velocity-area method is the most widely used method at present. This method calculates the flow by measuring the cross-sectional velocity and cross-sectional area. In the velocity-area method, radar flowmeters that integrate radar water level meters and current meters are widely used in online flow monitoring. Therefore, how to improve the data measurement accuracy of radar flowmeters is currently a relatively large problem.

[0004] The measurement of water level and flow rate by radar is actually the measurement of distance and speed, which can be regarded as a spectrum estimation problem. The measurement error of radar flowmeter mainly comes from the presence of clutter signals caused by environmental factors in the echo signal received by the radar. So how to eliminate the influence of clutter signals on the received water level and flow rate information and measure the parameter information more accurately? We need to add signal processing algorithms to the processing of echo signals and perform spectrum estimation on the signal spectrum. Since the measurement principles of water level and flow rate using radar technology are different, the selected band Fourier transform (Zoom-FFT) is used when measuring water level information, and the Burg algorithm of the AR model is used when measuring flow rate information, so as to improve the flow measurement accuracy of radar flowmeter. Summary of the invention

[0005] Purpose of the invention: To address the problem of flow meter measurement data accuracy in existing channel flow measurement technologies, this paper invents a channel flow measurement method based on the Burg algorithm of Zoom-FFT and AR model.

[0006] Technical solution: The present invention provides a channel flow measurement method based on the Burg algorithm of Zoom-FFT and AR model.

[0007] A channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model provided by the present invention includes:

[0008] Band-selected Fourier transform (Zoom-FFT): Zoom-FFT is a method that can perform local spectrum refinement, that is, reduce the interval between spectrum lines within the spectrum range of interest and improve the accuracy of spectrum analysis. The Zoom-FFT algorithm can greatly reduce the amount of calculation and improve the calculation speed. When the number of points in the original sequence is the same, the frequency resolution can be improved by using the Zoom-FFT algorithm. The basic idea of ​​the Zoom-FFT algorithm is to try to change the frequency domain coordinate position, move it to the left or right, so as to move the effective frequency band of the narrowband signal to the zero frequency position, and then analyze the spectrum. The Zoom-FFT algorithm can realize the amplification of the spectrum in a certain frequency band, so that its frequency resolution in this frequency band is improved and amplified by D times. Subsequently, because the infinite increase of the refinement multiple of the Zoom-FFT algorithm will lead to a decrease in accuracy, the energy center of gravity correction algorithm is used to correct the results obtained by the Zoom-FFT algorithm to further improve the ranging accuracy.

[0009] Burg algorithm for AR model: An autoregressive (AR) model can be viewed as an input sequence u(n) stimulating the output x(n) of a linear system H(z), where the input sequence u(n) is a white noise sequence with a variance of σ 2 The current output of the model is the weighted sum of the current input and the past p outputs. The output x(n) can be expressed as formula (1), where a k is the parameter of the AR model, and p is the order of the AR model.

[0010]

[0011] Power spectrum P x (e jw The relationship between ) and the AR model parameters is shown in formula (2).

[0012]

[0013] The Burg algorithm is one of the solutions to the AR model. The Burg algorithm can effectively estimate the required two small interval frequencies, and the model order will not be too large, and the amount of calculation is small. The Burg method uses a recursive method to first calculate the forward and backward prediction errors when the AR model order p ranges from 1 to m. The reflection coefficient k is obtained by using the forward and backward prediction errors of each order m, and then use the Levinson relation to obtain the parameters of the AR model, and substitute the parameters into equation (2) to obtain the power spectrum. The recursive formula for calculating the m-order prediction error is shown in equations (3) to (5), where m = 1, 2, ..., p.

[0014]

[0015]

[0016]

[0017] The formula for obtaining the reflection coefficient is as follows:

[0018]

[0019] Estimate k m After that, the parameters of the AR model at order m are still obtained recursively by the Levinson algorithm.

[0020] a m (k) = a m-1 (k)+ka m-1 (mk) (7)

[0021] a m (m) = k m (8)

[0022]

[0023] After obtaining the parameters of the model, the power spectrum can be estimated according to equation (2). Through the derivation of equations (3) to (9), the Burg algorithm not only saves a lot of measurement time, but also cleverly avoids making artificial assumptions about data outside the known data segment by using a recursive method.

[0024] Preferably, the channel flow measurement method comprises:

[0025] Component A1: Radar transceiver transmits frequency modulated continuous wave. Frequency modulated continuous wave radar transmits a continuous wave with a changing frequency during the frequency scanning cycle. The frequency of the reflected echo is different from the frequency of the source transmission signal, and an intermediate frequency signal with a frequency difference between the received wave and the transmitted wave is obtained.

[0026] Component A2: The water level meter module performs a band-selective Fourier transform (Zoom-FFT) on the intermediate frequency signal, performs spectrum refinement with a smaller refinement multiple, and uses the energy center of gravity correction algorithm to accurately estimate the intermediate frequency signal frequency. The basic idea of ​​the Zoom-FFT algorithm is: by locally refining the frequency of the signal, reducing the spectrum line interval within the spectrum range of interest, moving the effective frequency band of the narrowband signal to the zero frequency position, and then analyzing the spectrum to achieve spectrum refinement; then use the energy center of gravity correction algorithm to further improve the measurement accuracy and improve the measurement and extraction of water level information.

[0027] Component A3: The velocity meter module uses the Burg algorithm of the AR model to estimate the spectrum of the intermediate frequency signal. The accuracy of the discrete Fourier transform velocity measurement depends on the resolution of the discrete spectrum. When calculating through the fast Fourier transform (FFT), there will be a "fence effect", that is, the frequency of other points outside the N sampling points cannot be calculated. By adding the Burg algorithm of the AR model on the basis of the fast Fourier transform (FFT), the influence of the "fence effect" can be effectively eliminated and the accuracy of the velocity measurement can be improved.

[0028] Component A4: Calculate the flow rate according to the velocity-area method. The velocity-area method is to calculate the flow rate data of the cross section by the cross-sectional velocity and cross-sectional area. Figure 1 shown.

[0029] Preferably, the radar water level meter uses a zoom-FFT algorithm:

[0030] Step B1: Collect the echo signal through the radar transceiver, mix the transmission wave with the echo to obtain the radar intermediate frequency signal, perform fast Fourier transform on it to obtain the spectrum, and obtain the target frequency band;

[0031] Step B2: Spectrum shifting: moving the center frequency of the target frequency band to the zero point of the frequency axis;

[0032] Step B3: low-pass filtering the frequency-shifted signal;

[0033] Step B4: sampling the signal with a coefficient of D;

[0034] Step B5: Perform Fourier transform spectrum analysis to perform Fourier transform on the extracted signal;

[0035] Step B6: Frequency adjustment, moving a frequency spectrum obtained in step B5 to the actual frequency prescription to obtain the refined real frequency spectrum.

[0036] Preferably, the energy center of gravity correction algorithm steps:

[0037] Step C1: Perform fast Fourier transform on the intermediate frequency signal to find the spectral line position corresponding to the maximum value of the power spectrum;

[0038] Step C2: Find the position corresponding to the maximum value of the power spectrum after refinement in step B4, and perform energy correction calculation.

[0039] Preferably, the Burg algorithm steps of the AR model adopted by the radar velocity meter are:

[0040] Step D1: Collect echo signals through the radar transceiver, add adaptive gain adjustment to the echo signals, preset the threshold of the sampling signal amplitude, compare the sampling signal amplitude with the threshold after each sampling, and appropriately increase or decrease the gain if the sampling signal is too large or too small to ensure that the sampling signal is within the appropriate amplitude range;

[0041] Step D2: After front-end processing, obtain the radar intermediate frequency signal; perform fast Fourier transform on it to obtain the spectrum and obtain the target frequency band;

[0042] Step D3: After digital sampling, the intermediate frequency signal is preprocessed by filtering, windowing, etc. to obtain a digital signal sequence to be processed;

[0043] Step D4: Use fast Fourier transform (FFT) to perform frequency domain analysis on the sequence and obtain the signal frequency using peak detection;

[0044] Step D5: Calculate the AR model parameters corresponding to the signal sequence using the Burg method to obtain the corresponding power spectrum estimation, calculate the signal frequency using peak detection and compare it with the result obtained in step D3 to obtain the flow velocity.

[0045] Preferably, the radar flowmeter flow calculation steps are:

[0046] The velocity-area method is a flow estimation algorithm that obtains cross-sectional flow data by calculating the cross-sectional flow velocity and cross-sectional area.

[0047] Step E1: After the radar current meter and the radar water level meter are measured, the instantaneous flow velocity V1 and the channel water level h can be obtained. The velocity coefficient after calibration is k, and the average flow velocity of the cross section at this time is formula (10):

[0048] V m =k*V1 (10)

[0049] Step E2: The cross-sectional area of ​​the river at this time can be expressed by A(h), and the formula for the flow rate Q of the river section at this time can be expressed as formula (11):

[0050] Q=V m *A(h)=k*V1*A(h) (11)

[0051] A channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model, the characteristics of which include:

[0052] The radar transceiver transmits a frequency modulated continuous wave. The frequency modulated continuous wave radar transmits a continuous wave with a changing frequency during the frequency scanning cycle. The frequency of the reflected echo is different from the frequency of the signal transmitted by the source, and an intermediate frequency signal with a frequency difference between the received wave and the transmitted wave is obtained.

[0053] The water level meter module performs a band-selective Fourier transform (Zoom-FFT) on the intermediate frequency signal. The basic idea of ​​the Zoom-FFT algorithm is to try to change the frequency domain coordinate position, shift it to the left or right, so as to move the effective frequency band of the narrowband signal to the zero frequency position, and then analyze the spectrum to achieve spectrum refinement and improve the measurement and extraction of water level information.

[0054] The velocity meter module uses the Burg algorithm of the AR model to estimate the spectrum of the intermediate frequency signal. By adding the Burg algorithm of the AR model on the basis of the fast Fourier transform (FFT), the influence of the "fence effect" can be effectively eliminated and the accuracy of the velocity measurement can be improved.

[0055] The flow rate is converted according to the velocity-area method model. The velocity-area method is to obtain the flow data of the cross section by calculating the cross-sectional velocity and cross-sectional area.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. Improve the accuracy of spectrum estimation. The Zoom-FFT algorithm and the Burg algorithm of the AR model can effectively improve the spectrum estimation of the intermediate frequency signal, effectively eliminate the influence of the "fence effect" during spectrum estimation, and improve the accuracy of flow velocity measurement.

[0058] 2. Use the energy center of gravity correction algorithm to correct the frequency obtained by the Zoom-FFT algorithm to further improve the measurement accuracy.

[0059] 3. The calculation amount of the velocity-area method is small. According to the velocity-area method model, the flow rate can be converted by simply passing the cross-sectional flow velocity and cross-sectional area to obtain the cross-sectional flow data. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the flow chart of the components of the present invention.

[0061] Figure 2 The figure is a schematic diagram of a band-selective Fourier transform process implemented in the present invention.

[0062] Figure 3 FIG. 4 is a waveform diagram of a frequency modulated continuous wave according to a specific embodiment of the present invention.

[0063] Figure 4 The figure is a schematic diagram of the working principle of a radar water level meter according to a specific embodiment of the present invention.

[0064] Figure 5 The figure is a schematic diagram of the working principle of a radar current meter according to a specific embodiment of the present invention.

[0065] Figure 6 It is a schematic diagram of the velocity-area method model of the present invention. DETAILED DESCRIPTION

[0066] The present invention is further described below in conjunction with the accompanying drawings and examples. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0067] According to a channel flow measurement method of Burg algorithm based on Zoom-FFT and AR model provided by the present invention, Figure 1 As shown, including:

[0068] Component A1: Radar transceiver transmits frequency modulated continuous wave: In channel flow calculation, frequency modulated continuous wave is widely used due to its high measurement accuracy. The key to frequency modulated continuous wave radar distance and speed measurement lies in the measurement of intermediate frequency signal frequency. Usually, spectrum analysis methods such as fast Fourier transform (FFT), wavelet analysis and maximum likelihood estimation can accurately estimate the frequency of single frequency signal. The frequency of the frequency modulated continuous wave transmission signal is linearly related to time, and the echo signal is the delayed signal after the transmission signal is reflected by the target. The transmission signal and the echo signal are passed through a mixer to obtain an intermediate frequency signal. The distance and speed information of the target can be obtained by analyzing the intermediate frequency signal. Figure 3 As shown, the solid line represents the frequency of the transmitted signal, the dotted line represents the frequency of the received signal, the time t is the period of the frequency sweep, the bandwidth B is the bandwidth of the frequency sweep, and the echo signal is delayed after the transmitted signal passes through the target. In the frequency change of the triangle, the distance between the rising edge and the falling edge can be measured. If there is no Doppler frequency, the frequency difference of the rising edge is equal to the measured value of the falling edge. For moving objects, the frequency difference of the rising edge and the falling edge is different. The speed can be measured by these two frequency differences. The frequency differences that need to be measured are denoted as Δf1 and Δf2, and the known frequency modulation slope is K r After low-pass filtering and amplification, in order to achieve fast Fourier transform (FFT) and detection of the beat frequency signal, the beat frequency signal is transmitted to the digital signal processor link. After certain calculations, the target data information is input to the control terminal for display. The frequency modulated continuous wave radar uses positive and negative frequency modulation slopes to eliminate the coupling between distance and speed, and then estimates the target speed. The frequency modulated continuous wave waveform is as follows Figure 3 shown.

[0069] Component A2: The water level meter module performs a band-selective Fourier transform (Zoom-FFT) on the intermediate frequency signal: The frequency modulated continuous wave radar uses non-contact measurement to measure the water level. The water level is calculated based on the measured delay between the transmitted wave and the echo. It is not affected by the density, viscosity, and humidity of the medium and has a high measurement accuracy. Figure 4 shown.

[0070] The frequency modulated continuous wave radar water level meter transmits frequency modulated electromagnetic waves to the water surface. The echo signal reflected by the water surface is received by the radar and mixed with the transmitted signal. The frequency of the intermediate frequency signal generated is proportional to the distance from the water surface. The transmitted signal in one modulation cycle can be expressed as formula (12):

[0071]

[0072] In formula (12): S T is the amplitude of the transmitted signal; T is the modulation period; f0 is the center frequency of the transmitted signal; B is the frequency sweep bandwidth of the transmitted signal; is the initial phase of the transmitted signal. After the round-trip delay, the intermediate frequency signal received by the radar and mixed with the transmitted signal is simplified and amplitude normalized, which can be expressed as formula (13):

[0073]

[0074] The intermediate frequency signal obtained by mixing the transmitted signal and the echo signal is given by formula (14):

[0075]

[0076] It can be seen that the frequency of the intermediate frequency signal is proportional to the distance from the water surface. Since the reflection of electromagnetic waves by the water surface can be approximated as mirror reflection, the intermediate frequency signal is a single-frequency signal under ideal conditions. In order to improve the ranging accuracy, a signal processing algorithm is required to accurately measure the frequency of the intermediate frequency signal. Considering factors such as computational efficiency and resolution, the Zoom-FFT method is mainly used for refinement in this paper, that is, to reduce the interval between spectrum lines within the spectrum range of interest and improve the accuracy of spectrum analysis. Subsequently, the energy center of gravity correction algorithm is used to correct the frequency obtained by the Zoom-FFT algorithm to further improve the measurement accuracy.

[0077] Component A3: The velocity meter module uses the Burg algorithm of the AR model to estimate the spectrum of the intermediate frequency signal: the velocity meter measures the flow velocity by analyzing the spectrum of the intermediate frequency signal to obtain the Doppler frequency shift, and obtains the flow velocity information based on the Doppler frequency shift. Figure 5As shown in the figure, the continuous wave radar transmits electromagnetic waves of a certain frequency at a certain inclination angle θ to illuminate the water surface to produce scattering. A part of the scattered waves is received by the radar as echoes. The scattered waves satisfy the following Bragg conditions, as shown in formula (15):

[0078]

[0079] In formula (15), λ0 is the wavelength of the scattered wave, and λ is the wavelength of the incident wave. The continuous wave signal emitted by the radar is a sinusoidal signal with a frequency of f0, as shown in formula (16):

[0080]

[0081] When the water surface flows and the radar is stationary, according to the Doppler principle, the signal frequency of the radar echo will change, as shown in formula (17):

[0082]

[0083] In formula (17): f d is the Doppler frequency. After the radar echo is processed by front-end processes such as mixing and filtering, the frequency shift of the intermediate frequency signal to be processed is obtained, as shown in formula (18):

[0084]

[0085] In formula (18): f c is the radar carrier frequency, v is the water surface velocity, c is the speed of light, and θ is the inclination angle of the radar irradiating the water surface. Transforming equation (18) yields equation (19):

[0086]

[0087] From formula (19), we can see that the carrier frequency f c , the speed of light c are known constants, and the inclination angle θ can be set and directly measured as needed, so it is only necessary to detect the Doppler signal f from the echo signal. d , the velocity of the water surface can be calculated. Therefore, the measurement of flow velocity can be regarded as a spectrum estimation problem. In order to effectively eliminate the influence of the "fence effect", the Burg algorithm of the AR model is added on the basis of the fast Fourier transform (FFT) to improve the accuracy of flow velocity measurement.

[0088] Component A4: Calculate the flow rate according to the velocity-area method model, such as Figure 6 The velocity-area method is to calculate the flow data of the cross section by the cross-sectional velocity and cross-sectional area.

[0089] Specifically, the radar water level meter uses a zoom-FFT algorithm, such as Figure 2 As shown, including:

[0090] Step B1: Collect the echo signal through the radar transceiver, obtain the radar intermediate frequency signal after front-end processing, perform fast Fourier transform on it to obtain the spectrum, and obtain the target frequency band;

[0091] Step B2: Spectrum shifting: moving the center frequency of the target frequency band to the zero point of the frequency axis;

[0092] Step B3: Low-pass filter the frequency-shifted signal to filter out the high-frequency components that are not of concern, because in the next step, the signal will be decimated by a coefficient of D, reducing the equivalent sampling rate to According to Nyquist sampling theorem, we must make All components are filtered out to prevent spectrum aliasing;

[0093] Step B4: Sample the signal with a coefficient of D; that is, keep one sampling point every D sampling points in the time domain and discard the rest. The equivalent sampling rate after decimation is

[0094] Step B5: Perform Fourier transform spectrum analysis and perform N-point Fourier transform on the extracted time domain signal. Since the equivalent sampling frequency of the signal at this time is Therefore, after Zoom-FFT processing, the spectral line interval in the selected spectrum range is Where Δf is the spectral line spacing when performing N-point FFT directly on the intermediate frequency signal;

[0095] Step B6: Frequency adjustment, moving a frequency spectrum obtained in step B5 to the actual frequency prescription to obtain the refined real frequency spectrum.

[0096] Specifically, the energy center of gravity correction algorithm includes:

[0097] Step C1: s After the intermediate frequency signal is subjected to N-point fast Fourier transform, the spectral line position corresponding to the maximum value of the power spectrum is found. The power spectrum is shown in formula (20):

[0098] P(k)=|X(k)| 2 (20)

[0099] Step C2: Find the position corresponding to the maximum value of the refined power spectrum in step B4 And perform energy correction calculation, as shown in formula (21):

[0100]

[0101] Then the liquid level distance is calculated as shown in formula (22):

[0102]

[0103] Specifically, the radar velocity meter uses the Burg algorithm of the AR model, including:

[0104] Step D1: Collect echo signals through the radar transceiver, add adaptive gain adjustment to the echo signals, preset the threshold of the sampling signal amplitude, compare the sampling signal amplitude with the threshold after each sampling, and appropriately increase or decrease the gain if the sampling signal is too large or too small to ensure that the sampling signal is within the appropriate amplitude range;

[0105] Step D2: After front-end processing, obtain the radar intermediate frequency signal; perform fast Fourier transform on it to obtain the spectrum and obtain the target frequency band;

[0106] Step D3: After digital sampling, the intermediate frequency signal is preprocessed by filtering, windowing, etc. to obtain a digital signal sequence to be processed;

[0107] Step D4: Use fast Fourier transform (FFT) to perform frequency domain analysis on the sequence and obtain the signal frequency using peak detection;

[0108] Step D5: Calculate the AR model parameters corresponding to the signal sequence using the Burg method to obtain the corresponding power spectrum estimation, calculate the signal frequency using peak detection and compare it with the result obtained in step D3 to obtain the flow velocity.

[0109] Specifically, for step D5, an autoregressive (AR) model can be viewed as an input sequence u(n) stimulating the output x(n) of a linear system H(z), where the input sequence u(n) is a white noise sequence with a variance of σ 2 The current output of the model is the weighted sum of the current input and the past p outputs. The output x(n) can be expressed as formula (1), where a k is the parameter of the AR model, and p is the order of the AR model.

[0110]

[0111] Power spectrum P x (e jw The relationship between ) and the AR model parameters is shown in formula (2).

[0112]

[0113] The Burg algorithm is one of the solutions to the AR model. The Burg algorithm can effectively estimate the required two small interval frequencies, and the model order will not be too large, and the amount of calculation is small. The Burg method uses a recursive method to first calculate the forward and backward prediction errors when the AR model order p ranges from 1 to m. The reflection coefficient k is obtained by using the forward and backward prediction errors of each order m , and then use the Levinson relation to obtain the parameters of the AR model, and substitute the parameters into equation (2) to obtain the power spectrum. The recursive formula for calculating the m-order prediction error is shown in equations (3) to (5), where m = 1, 2, ..., p.

[0114]

[0115]

[0116]

[0117] The formula for obtaining the reflection coefficient is as follows:

[0118]

[0119] Estimate k m After that, the parameters of the AR model at order m are still obtained recursively by the Levinson algorithm.

[0120] a m (k) = a m-1 (k)+ka m-1 (mk) (7)

[0121] a m (m) = k m (8)

[0122]

[0123] After obtaining the parameters of the model, the power spectrum can be estimated according to equation (2). Through the derivation of equations (3) to (9), the Burg algorithm not only saves a lot of measurement time, but also cleverly avoids making artificial assumptions about data outside the known data segment by using a recursive method.

[0124] Specifically, the radar flow meter calculation steps are:

[0125] The velocity-area method is a flow estimation algorithm that obtains cross-sectional flow data by calculating the cross-sectional flow velocity and cross-sectional area.

[0126] Step E1: After the radar current meter and the radar water level meter are measured, the instantaneous flow velocity V1 and the channel water level h can be obtained. The velocity coefficient after calibration is k, and the average flow velocity of the cross section at this time is formula (10):

[0127] V m =k*V1 (10)

[0128] Step E2: The cross-sectional area of ​​the river at this time can be expressed by A(h), and the formula for the flow rate Q of the river section at this time can be expressed as formula (11):

[0129] Q=V m *A(h)=k*V1*A(h) (11)

[0130] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model Component A1: Radar transceiver transmits frequency modulated continuous wave. Frequency modulated continuous wave radar transmits a continuous wave with a changing frequency during the frequency scanning cycle. The frequency of the reflected echo is different from the frequency of the source transmission signal. After mixing the transmission wave with the echo, an intermediate frequency signal with the frequency difference between the received wave and the transmission wave is obtained. Component A2: The water level meter module performs a band-selective Fourier transform (Zoom-FFT) on the intermediate frequency signal, performs spectrum refinement with a smaller refinement multiple, and uses the energy center of gravity correction algorithm to accurately estimate the intermediate frequency signal frequency. The basic idea of ​​the Zoom-FFT algorithm is: by locally refining the frequency of the signal, reducing the spectrum line interval within the spectrum range of interest, moving the effective frequency band of the narrowband signal to the zero frequency position, and then analyzing the spectrum to achieve spectrum refinement; then use the energy center of gravity correction algorithm to further improve the measurement accuracy and improve the measurement and extraction of water level information. Component A3: The velocity meter module uses the Burg algorithm of the AR model to estimate the spectrum of the intermediate frequency signal. The accuracy of the discrete Fourier transform velocity measurement depends on the resolution of the discrete spectrum. When calculating through the fast Fourier transform (FFT), there will be a "fence effect", that is, the frequency of other points outside the N sampling points cannot be calculated. By adding the Burg algorithm of the AR model on the basis of the fast Fourier transform (FFT), the influence of the "fence effect" can be effectively eliminated and the accuracy of the velocity measurement can be improved. Component A4: Calculate flow rate according to the velocity-area method model. The velocity-area method is to calculate the flow data of the cross section by the cross-sectional velocity and cross-sectional area.

2. According to claim 1, a channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model is characterized in that: Zoom-FFT: Step B1: Collect the echo signal through the radar transceiver, mix the transmission wave with the echo to obtain the radar intermediate frequency signal, perform fast Fourier transform on it to obtain the spectrum, and obtain the target frequency band; Step B2: Spectrum shifting: moving the center frequency of the target frequency band to the zero point of the frequency axis; Step B3: low-pass filtering the frequency-shifted signal; Step B4: sampling the signal with a coefficient of D; Step B5: Perform Fourier transform spectrum analysis to perform Fourier transform on the extracted signal; Step B6: Frequency adjustment, moving a frequency spectrum obtained in step B5 to the actual frequency prescription to obtain the refined real frequency spectrum.

3. The channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model according to claim 1 is characterized in that: Energy center of gravity correction algorithm: Step C1: Perform fast Fourier transform on the intermediate frequency signal to find the spectral line position corresponding to the maximum value of the power spectrum; Step C2: Find the position corresponding to the maximum value of the power spectrum after refinement in step B4, and perform energy correction calculation.

4. The channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model according to claim 1 is characterized in that: Burg algorithm based on AR model: Step D1: Collect echo signals through the radar transceiver, add adaptive gain adjustment to the echo signals, preset the threshold of the sampling signal amplitude, compare the sampling signal amplitude with the threshold after each sampling, and appropriately increase or decrease the gain if the sampling signal is too large or too small to ensure that the sampling signal is within the appropriate amplitude range; Step D2: After front-end processing, obtain the radar intermediate frequency signal; perform fast Fourier transform on it to obtain the spectrum and obtain the target frequency band; Step D3: After digital sampling, the intermediate frequency signal is preprocessed by filtering and windowing to obtain a digital signal sequence to be processed; Step D4: Use fast Fourier transform (FFT) to perform frequency domain analysis on the sequence and obtain the signal frequency using peak detection; Step D5: Calculate the AR model parameters corresponding to the signal sequence using the Burg method to obtain the corresponding power spectrum estimation, calculate the signal frequency using peak detection and compare it with the result obtained in step D3 to obtain the flow velocity.

5. The channel flow measurement method based on Burg algorithm of Zoom-FFT and AR model according to claim 1, wherein the characteristics include: The radar transceiver transmits a frequency modulated continuous wave. The frequency modulated continuous wave radar transmits a continuous wave with a changing frequency during the frequency scanning cycle. The frequency of the reflected echo is different from the frequency of the source transmission signal. After mixing the transmission wave with the echo, an intermediate frequency signal with a frequency difference between the received wave and the transmission wave is obtained. The water level meter module performs a band-selective Fourier transform (Zoom-FFT) on the intermediate frequency signal, performs spectrum refinement with a smaller refinement multiple, and uses the energy center of gravity correction algorithm to accurately estimate the intermediate frequency signal frequency. The basic idea of ​​the Zoom-FFT algorithm is: by locally refining the frequency of the signal, reducing the spectrum line interval within the spectrum range of interest, moving the effective frequency band of the narrowband signal to the zero frequency position, and then analyzing the spectrum to achieve spectrum refinement; then use the energy center of gravity correction algorithm to further improve the measurement accuracy and improve the measurement and extraction of water level information. The velocity meter module uses the Burg algorithm of the AR model to estimate the spectrum of the intermediate frequency signal. By adding the Burg algorithm of the AR model on the basis of the fast Fourier transform (FFT), the influence of the "fence effect" can be effectively eliminated and the accuracy of the velocity measurement can be improved. The flow rate is converted according to the velocity-area method model. The velocity-area method is to obtain the flow data of the cross section by calculating the cross-sectional velocity and cross-sectional area.

Images