Gas flow measurement error correction method and system based on critical flow nozzle

Abstract

This invention discloses a method and system for correcting gas flow measurement errors based on a critical flow sonic nozzle, relating to the field of gas flow metering technology. The method includes performing logarithmic operations on the sharpened average power spectrum and executing an inverse discrete Fourier transform to obtain a real cepstrum sequence. By extracting features from the low cepstrum frequency range, characteristic factors representing the nozzle throat boundary layer state are calculated. Simultaneously, the Reynolds number under the current flow condition is calculated based on static pressure and static temperature data, and the ISO standard outflow coefficient is dynamically corrected using a pre-calibrated Reynolds number partition sensitivity coefficient to obtain the corrected outflow coefficient. By performing autocorrelation analysis on the high time delay range of the real cepstrum, the critical flow state of the nozzle is determined. Under stable critical flow conditions, the actual mass flow rate is calculated based on the corrected outflow coefficient, improving the accuracy and stability of flow measurement under wide operating conditions and reducing the systematic errors introduced by traditional static correction models.

Description

Technical Field

[0001] This invention relates to the field of gas flow measurement technology, and in particular to a method and system for correcting gas flow measurement errors based on a critical flow sonic nozzle. Background Technology

[0002] Critical flow sonic nozzles, as high-precision gas flow metering devices based on compressible fluid dynamics theory, are widely used in gas flow metering and traceability, standard device calibration, and industrial process control because their mass flow rate under critical flow conditions is only related to the upstream thermal state and nozzle geometric parameters and is not affected by downstream pressure fluctuations.

[0003] Existing gas flow measurement technologies based on critical flow sonic nozzles still have certain shortcomings in terms of high accuracy and long-term stability. The ISO standard model and its extended methods usually equate factors such as nozzle throat boundary layer effects, surface roughness evolution, and micro-structural changes to static or weakly changing empirical correction terms, which are difficult to reflect the dynamic characteristics of boundary layer state changes with operating conditions during actual operation. Especially in high Reynolds number or wide operating range, such simplification is prone to introducing systematic errors. Summary of the Invention

[0004] In view of the problems existing in the prior art, the present invention is proposed.

[0005] Therefore, this invention provides a method and system for correcting gas flow measurement errors based on a critical flow sonic nozzle, in order to solve the problem that ISO standard models and their extended methods usually equate factors such as nozzle throat boundary layer effects, surface roughness evolution, and minute structural changes to static or weakly varying empirical correction terms, which are difficult to reflect the dynamic characteristics of boundary layer state changes with operating conditions during actual operation. Especially in high Reynolds numbers or wide operating ranges, such simplification is prone to introducing systematic errors.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for correcting gas flow measurement errors based on a critical flow sonic nozzle, comprising the following steps:

[0008] Multi-source data is collected and preprocessed. The AC pulsation sequence is segmented and multiplied point by point with the Hamming window function to obtain the windowed signal sequence. The windowed signal sequence is then subjected to wavelet transform to determine the wavelet coefficients falling within the frequency axis neighborhood. The average energy along the time dimension is used to obtain the sharpened average power spectrum.

[0009] Apply the natural logarithm to the sharpened average power spectrum and perform the inverse discrete Fourier transform to obtain the real cepstral sequence. Calculate the boundary layer characteristic factor, calculate the Reynolds number under the current flow condition, and convert it into the corrected outflow coefficient.

[0010] The autocorrelation function of the real cepstrum in the high time delay interval is calculated, and the state of the critical flow sonic nozzle is judged. Error correction is performed based on the judgment results.

[0011] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle described in this invention, the step of segmenting the AC pulsating sequence and multiplying it point-by-point with a Hamming window function to obtain a windowed signal sequence, performing wavelet transform on the windowed signal sequence to determine the wavelet coefficients falling within the frequency axis neighborhood, and obtaining the sharpened average power spectrum by averaging the energy along the time dimension, includes:

[0012] The AC pulsation sequence is segmented using the sliding window method to obtain a rectangular window sequence. The Hamming window function is then multiplied point by point with the rectangular window sequence to obtain the windowed signal sequence.

[0013] Perform continuous wavelet transform on the windowed signal sequence to obtain wavelet coefficients, determine the instantaneous angular frequency by the partial derivative of the phase of the wavelet coefficients, and convert it into an instantaneous frequency;

[0014] Define a frequency axis, aggregate all wavelet coefficients whose instantaneous frequencies fall within the neighborhood of the frequency axis, and obtain the time-frequency coefficients after synchronous squeezing transformation;

[0015] The average energy of the time-frequency coefficients along the time dimension is calculated to obtain the sharpened average power spectrum.

[0016] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle described in this invention, the step of applying a natural logarithm operation to the sharpened average power spectrum and performing an inverse discrete Fourier transform to obtain a real cepstral sequence, and calculating the boundary layer characteristic factor, includes:

[0017] Apply the natural logarithm to each element of the sharpened average power spectrum to obtain the logarithmic power spectrum. Perform an inverse discrete Fourier transform on the logarithmic power spectrum to obtain the real cepstrum sequence.

[0018] Set a time delay threshold, extract the low time delay components from the inverted frequency index from 1 to the time delay threshold in the real cepstrum, calculate the sum of squares, and take the square root to obtain the boundary layer feature factor.

[0019] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle described in this invention, the step of calculating the Reynolds number under the current flow condition and converting it into a corrected discharge coefficient includes:

[0020] Calculate the Reynolds number under the current flow conditions based on static pressure and static temperature;

[0021] Collect the historical Reynolds number of the clean nozzle and calculate the mean as the global feature reference value of the clean nozzle. Subtract the global feature reference value from the boundary layer feature factor to obtain the boundary layer feature deviation.

[0022] The entire operating Reynolds number range of the nozzle is divided into continuous sub-intervals at equal intervals. The midpoint value of the sub-interval is calculated as the calibration Reynolds number. The compensation coefficient is calculated by combining the boundary layer characteristic deviation and the calibration Reynolds number.

[0023] Based on the current Reynolds number, the baseline outflow coefficient is determined using the ISO 9300 standard, and multiplied by the current compensation coefficient to obtain the corrected outflow coefficient.

[0024] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle described in this invention, the step of calculating the autocorrelation function of the real cepstral spectrum in the high time delay interval and determining the state of the critical flow sonic nozzle includes:

[0025] Preset a high time delay interval and calculate the autocorrelation function of the real cepstrum in the high time delay interval;

[0026] Set a judgment threshold, and compare the maximum value of the autocorrelation function with the judgment threshold. If the maximum value of the autocorrelation function is greater than the judgment threshold, it is judged as a non-critical flow state; otherwise, it is judged as a stable critical flow state.

[0027] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle according to the present invention, the error correction based on the judgment result includes:

[0028] When the flow state is a steady critical flow state, the actual mass flow rate of the nozzle is calculated based on the corrected discharge coefficient.

[0029] When the flow state is noncritical, a noncritical flow warning signal is issued, and the actual mass flow rate calculated in the previous stable critical flow state is output.

[0030] As a preferred embodiment of the gas flow measurement error correction method based on a critical flow sonic nozzle described in this invention, the step of acquiring multi-source data and performing preprocessing includes:

[0031] Upstream of the sonic nozzle, smart sensors are used to collect multi-source data and perform preprocessing.

[0032] The intelligent sensors include dynamic pressure, absolute pressure, and temperature sensors;

[0033] The multi-source data includes voltage signals, static pressure, and static temperature data;

[0034] The preprocessing includes converting the voltage signal value into the corresponding instantaneous absolute pressure, splicing them into a pressure sequence in chronological order, applying a first-order recursive digital high-pass filter to the pressure sequence to obtain an AC pulsation sequence.

[0035] The static pressure and static temperature are denoised and standardized.

[0036] Secondly, the present invention provides a gas flow measurement error correction system based on a critical flow sonic nozzle, comprising:

[0037] The acquisition signal preprocessing module is used to acquire dynamic pressure, static pressure and static temperature multi-source data upstream of the sonic nozzle, and convert the voltage signal into a pressure pulsation sequence to complete noise reduction and standardization processing.

[0038] The boundary layer feature separation module is used to perform time-frequency analysis on the pressure pulsation signal through sliding window, wavelet transform and synchronous extrusion redistribution to obtain a sharpened average power spectrum. Based on cepstral analysis, it separates the turbulence source characteristics from the nozzle boundary layer system characteristics and extracts feature factors that reflect the throat boundary layer state.

[0039] The calibration and dynamic compensation module is used to combine the real-time Reynolds number with the pre-stored sensitivity coefficient to construct a boundary layer influence compensation model and calculate the corrected real-time outflow coefficient.

[0040] The critical flow discrimination and correction module is used to determine whether the nozzle is in a stable critical flow state, and outputs a high-precision mass flow rate based on the corrected discharge coefficient when the condition is met; otherwise, it issues an alarm.

[0041] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the gas flow measurement error correction method based on a critical flow sonic nozzle as described in the first aspect of the present invention.

[0042] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the gas flow measurement error correction method based on a critical flow sonic nozzle as described in the first aspect of the present invention.

[0043] The beneficial effects of this invention are as follows: By combining multi-source dynamic pressure sensing with synchronous extrusion time-frequency analysis, and with cepstral domain boundary layer feature separation and Reynolds number partition compensation modeling, this invention improves the dynamics and precision of the nozzle throat boundary layer state characterization, thereby enhancing the accuracy and long-term stability of gas flow measurement under wide operating conditions and high Reynolds number conditions. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0045] Figure 1 This is a flowchart illustrating the operation of the gas flow measurement error correction method based on a critical flow sonic nozzle in Example 1.

[0046] Figure 2 This is a schematic diagram of the gas flow measurement error correction system based on a critical flow sonic nozzle in Example 1. Detailed Implementation

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0048] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for correcting gas flow measurement errors based on a critical flow sonic nozzle, including the following steps:

[0051] S1. Collect multi-source data and preprocess it. Segment the AC pulsation sequence and multiply it point by point with the Hamming window function to obtain the windowed signal sequence. Perform wavelet transform on the windowed signal sequence to determine the wavelet coefficients falling in the neighborhood of the frequency axis and the average energy along the time dimension to obtain the sharpened average power spectrum.

[0052] Specifically, this involves collecting and preprocessing multi-source data, including:

[0053] A dynamic pressure sensor is configured in the flow region upstream of the sonic nozzle to collect voltage signals, and an absolute pressure sensor and a temperature sensor are configured in the stable section upstream of the sonic nozzle to collect static pressure and static temperature, and perform preprocessing.

[0054] The preprocessing includes using a linear scaling transformation formula to convert the voltage signal value into the corresponding instantaneous absolute pressure, and then concatenating them into a pressure sequence in chronological order. The formula is as follows:

[0055] ,

[0056] in This represents the instantaneous absolute pressure value corresponding to the nth sampling point. The voltage signal at the nth sampling point output by the sensor. S is the output voltage of the sensor at zero pressure, and S is the sensitivity of the pressure sensor, which is obtained from the calibration certificate included with the product.

[0057] A first-order recursive digital high-pass filter is applied to the pressure sequence to obtain an AC pulsation sequence;

[0058] The static pressure and static temperature are denoised and standardized.

[0059] After linear scaling and high-pass filtering, the resulting AC pulsation sequence of the dynamic pressure signal only reflects the rapidly changing disturbance components in the flow field. Through first-order recursive digital high-pass filtering, static pressure and slowly changing components are systematically removed, allowing subsequent wavelet analysis to focus on high-frequency pulsation information that is truly related to jet instability, acoustic oscillation, or vortex structure evolution, thereby reducing pseudo-frequency components.

[0060] Furthermore, the AC pulsation sequence is segmented and multiplied point-by-point with a Hamming window function to obtain a windowed signal sequence. The windowed signal sequence is then subjected to wavelet transform to determine the wavelet coefficients falling within the frequency axis neighborhood. The average energy along the time dimension is used to obtain the sharpened average power spectrum, including:

[0061] The AC pulsation sequence is segmented using the sliding window method to obtain a rectangular window sequence, as shown in the formula:

[0062] ,

[0063] ,

[0064] in Let u be the sequence of rectangular windows, where u is the window index. This represents the sliding step size (number of sampling points). Let m be the starting index of the u-th window, and m be the local sample index within the window (0 to L-1, where L is the window length). This refers to the value of the m-th sampling point in the u-th window of the AC pulse sequence.

[0065] The windowed signal sequence is obtained by multiplying the Hamming window function with the rectangular window sequence point by point.

[0066] To analyze the changes in the local frequency characteristics of the windowed signal sequence over time, a continuous wavelet transform is performed on it;

[0067] Performing a continuous wavelet transform on the windowed signal sequence yields the wavelet coefficients of the signal in the two-dimensional planes of scale (corresponding frequency) and time. The Morlet wavelet is chosen as the mother wavelet function because it exhibits good localization properties in both the time and frequency domains, making it suitable for analyzing oscillating signals. The Morlet wavelet is a Gaussian window function modulated by a complex exponential function, as shown in the following formula:

[0068] ,

[0069] in To be at discrete scale and discrete time translation wavelet coefficients at the location, Let j be the discrete scaling factor. Let k be the discrete-time shift factor. For the windowed signal sequence at the m-th sampling point, W(a,b) is the complex conjugate of the Morlet mother wavelet function, a two-dimensional complex wavelet coefficient matrix, where the rows correspond to the scale index j (frequency) and the columns correspond to the time shift index k (time).

[0070] According to analytical wavelet theory, the instantaneous angular frequency is determined by the partial derivatives of the phase of the wavelet coefficients, and then converted into the instantaneous frequency using the following formula:

[0071] ,

[0072] ,

[0073] in To be at discrete scale and discrete time translation The instantaneous angular frequency determined at a given location, To take the imaginary part of a complex number, The partial derivatives of the wavelet coefficients with respect to the continuous-time shift variable b are approximated by the central difference along the time index k. The value at that location, The conversion coefficient between angular frequency and frequency. To be at discrete scale and discrete time translation The instantaneous frequency determined at a given location;

[0074] Each wavelet coefficient is converted from the original scale-time coordinate. Remapping to frequency-time coordinates superior;

[0075] The frequency axis is defined using a linear division method, with the following formula:

[0076] ,

[0077] in To redistribute the l-th frequency value on the target frequency axis, The lowest frequency for analysis, is the uniform interval of the target frequency axis, and l is the index of the target frequency axis;

[0078] By aggregating all instantaneous frequencies and determining the wavelet coefficients falling within the frequency axis neighborhood, the time-frequency coefficients after synchronous squeezing transform are obtained, as shown in the formula:

[0079] ,

[0080] in For the frequency axis and time The time-frequency coefficients after synchronous compression transformation at the location It is an energy compensation factor used to make the redistributed energy representation consistent with the Fourier spectrum of the signal in meaning;

[0081] Calculate the energy (square of the modulus) of the time-frequency coefficient at each time-frequency point, and then average it over all time points to obtain a one-dimensional frequency function, i.e., sharpen the average power spectrum.

[0082] The Hamming window attenuates smoothly at the window edge, effectively suppressing energy spillover caused by non-periodic signal truncation. The sliding window method, while maintaining continuous-time evolution characteristics, introduces an overlap analysis mechanism, enabling the system to capture the transient evolution of non-stationary oscillation structures in the jet, avoiding the limitation of traditional whole-segment Fourier analysis which can only obtain global average characteristics. The Morlet wavelet has both good time and frequency localization characteristics, which can accurately characterize the modulation oscillations, intermittent unstable structures, and multimodal coexistence phenomena in the jet. Time averaging can effectively suppress transient random disturbances, making the spectrum results more stable. Compared with directly averaging the Fourier spectrum, the sharpened average power spectrum of this invention can still accurately reflect the dominant frequency structure under non-stationary conditions, avoiding the problems of spectrum broadening and dominant frequency shift.

[0083] S2. Apply the natural logarithm to the sharpened average power spectrum and perform the inverse discrete Fourier transform to obtain the real cepstral sequence. Calculate the boundary layer characteristic factor, calculate the Reynolds number under the current flow condition, and convert it into the corrected outflow coefficient.

[0084] Specifically, the natural logarithm is applied to the sharpened average power spectrum, and an inverse discrete Fourier transform is performed to obtain the real cepstral sequence. Boundary layer feature factors are then calculated, including:

[0085] To transform the product relationship between the "source" and "system" in the frequency domain of the convolutional mixture into an additive relationship, the natural logarithm is applied to each element of the sharpened average power spectrum to obtain the logarithmic power spectrum, as shown in the formula:

[0086] ,

[0087] in For discrete frequency points At that point, the calculated logarithmic power spectral density value, It is the natural logarithm function. For discrete frequency points Sharpened average power spectrum at;

[0088] According to linear system theory, if the collected pressure pulsation signal The model is the convolution of the upstream background turbulence source signal s(t) and the impulse response h(t) of the nozzle throat acoustic system (whose characteristics are mainly determined by the boundary layer impedance), i.e. ,in For convolution operations, the power spectral density correspondingly satisfies , This represents the power spectral density of the source (turbulence). The power transfer function of the system (boundary layer) is represented by the power transfer function. To obtain the power density distribution in the frequency domain, taking the natural logarithm of the power spectrum transforms the above multiplicative relationship into an additive one. Sharpening the average power spectrum is... High-resolution estimation;

[0089] Perform an inverse discrete Fourier transform on the logarithmic power spectrum to convert it from the frequency domain to the "inverted frequency" domain, resulting in a real cepstrum sequence.

[0090] In cepstral analysis, this operation transforms the logarithmic power spectrum in the frequency domain to the so-called "inverted frequency" domain. The result is called the real cepstral spectrum, whose horizontal axis has the dimension of time. Its low inverted frequency components correspond to slowly changing components in the signal (in this case, system characteristics such as boundary layer impedance), while the high inverted frequency components correspond to rapidly changing components (source characteristics such as turbulent fluctuations).

[0091] Set a time delay threshold, extract the low-delay components from the real cepstral frequency index from 1 to the time delay threshold, calculate the sum of squares, and take the square root to obtain the boundary layer feature factor. The formula is as follows:

[0092] ,

[0093] ,

[0094] in For the time delay threshold, Where is the sampling frequency of the dynamic pressure sensor, and D is the diameter of the sonic nozzle throat. The speed of sound of the gas under the current operating conditions. It is the specific heat ratio of the gas being measured (the ratio of the specific heat capacity at constant pressure to the specific heat capacity at constant volume). and This is static pressure and static temperature data.

[0095] By applying the natural logarithm to the sharpened average power spectrum, the turbulent excitation source characteristics and the nozzle throat boundary layer acoustic system characteristics, which were originally multiplicative in the frequency domain, are transformed into an additive superposition relationship. This makes the system's transmission characteristics "explicit" in mathematical expression. The low reciprocal frequency components mainly reflect the system response characteristics that change slowly with frequency, while the high reciprocal frequency components mainly correspond to the rapidly changing turbulent excitation components. The time delay threshold is jointly determined by the nozzle throat diameter and the current sound velocity, so that the extracted boundary layer feature factors are naturally bound to the specific nozzle structure and gas state, avoiding the lack of universality caused by empirical thresholds. By focusing the energy of the low reciprocal frequency components, the influence of random noise and high-frequency turbulent fluctuations on the results is effectively suppressed, giving the feature factors good stability under repeated experiments and long-term operation conditions.

[0096] Furthermore, the Reynolds number under the current flow condition is calculated and converted into the corrected outflow coefficient, including:

[0097] The Reynolds number under the current flow condition is calculated based on static pressure and static temperature using the following formula:

[0098] ,

[0099] ,

[0100] in The Reynolds number under the current flow condition. For upstream gas density, Let be the dynamic viscosity of the gas at the current temperature, calculated using the Sutherland formula. The specific gas constant of the gas being measured is obtained by dividing the universal gas constant by the molar mass of the gas.

[0101] Collect the historical Reynolds number of the clean nozzle and calculate the mean as the global feature reference value of the clean nozzle. Subtract the global feature reference value from the boundary layer feature factor to obtain the boundary layer feature deviation.

[0102] The entire operating Reynolds number range of the nozzle is divided into three consecutive sub-intervals at equal intervals. For example, three Reynolds number intervals are divided: low, medium, and high. A standard roughness throat replacement section is prepared. The shape and installation dimensions of the replacement section are exactly the same as the original nozzle throat, but its inner wall has been sandblasted or otherwise standardized to have a known and uniform absolute roughness. The roughness value is certified by a metrology institution. During calibration, this replacement section is installed in the nozzle.

[0103] Calculate the midpoint value of the subinterval as the calibration Reynolds number, and conduct two experiments.

[0104] Two experiments were conducted, including the use of a pristine, clean nozzle throat. After the flow stabilized, the actual mass flow rate measured by a standard flow meter was recorded, and the boundary layer characteristic factor at this point was calculated.

[0105] Using a standard roughness throat replacement section, under the same calibrated Reynolds number conditions, after the flow stabilizes, record the actual mass flow rate measured by the standard flow meter and calculate the boundary layer characteristic factor at this time.

[0106] The sensitivity coefficient, converted to an interval, is calculated using the following formula:

[0107] ,

[0108] in Let be the boundary layer influence coefficient for the i-th Reynolds number interval. and This refers to the actual mass flow rate measured by a standard flow meter (such as a turbine flow meter or Coriolis mass flow meter) under operating conditions, when using a clean nozzle and when replacing the rough throat section. The theoretical mass flow rate is calculated based on global characteristic reference values ​​and the ISO 9300 standard formula. and The boundary layer characteristic factor values ​​are for the clean nozzle and rough throat replacement stages under calibrated Reynolds number conditions.

[0109] Compare the current real-time Reynolds number with the boundaries of each interval to determine the pre-stored sensitivity coefficient corresponding to the interval index;

[0110] The compensation coefficient is calculated using the following formula:

[0111] ,

[0112] in This is the current compensation coefficient. This refers to the pre-stored sensitivity coefficient corresponding to the interval where the current Reynolds number is located. This represents the current boundary layer feature deviation.

[0113] Based on the current Reynolds number, the baseline outflow coefficient determined using the ISO 9300 standard is multiplied by the current compensation coefficient to obtain the corrected outflow coefficient.

[0114] The influence of the boundary layer on the discharge coefficient varies significantly across different Reynolds number ranges. By using partitioned modeling, the problem of insufficient applicability of a single correction coefficient across the entire operating range can be avoided. By introducing comparative experiments with clean nozzles and standard roughness throat replacement sections at the calibrated Reynolds number points, the actual flow measurement results are combined with the ISO9300 theoretical model, enabling the sensitivity coefficient to have both experimental reliability and theoretical traceability. It no longer relies on periodic disassembly and inspection or manual experience judgment, and can reflect the changes in the boundary layer state of the nozzle throat in real time based on the pressure pulsation signal.

[0115] S3. Calculate the autocorrelation function of the real cepstrum in the high time delay interval, and make a state judgment on the critical flow sonic nozzle, and make error correction based on the judgment result;

[0116] Specifically, the autocorrelation function of the real cepstrum in the high time delay interval is calculated, and the state of the critical flow sonic nozzle is determined, including:

[0117] Based on empirical rules, a high-delay interval is predefined, and the autocorrelation function of the real cepstrum in the high-delay interval is calculated. The formula is as follows:

[0118] ,

[0119] in Let be a sequence of autocorrelation functions, representing the th... One autocorrelation function, Indicates the number of lag points. and For the termination delay point and the start delay point, The term "real cepstral" indicates that the sharpened real cepstral sequence is time-delayed. The amplitude at that point, The summation index variable represents the position of the time delay point in the real cepstral;

[0120] Use 3 The rule sets a judgment threshold and compares the maximum value of the autocorrelation function with the judgment threshold. If the maximum value of the autocorrelation function is greater than the judgment threshold, it is judged as a non-critical flow state with periodic disturbances; otherwise, it is judged as a stable critical flow state.

[0121] When a significant autocorrelation peak is detected in the high time delay interval, even if the pressure ratio surface meets the critical flow condition, the present invention can still determine it as a non-critical flow state, thereby avoiding the output of incorrect mass flow rate results under unreliable operating conditions.

[0122] Furthermore, error correction is performed based on the judgment results, including:

[0123] When the flow state is a steady critical flow state, the actual mass flow rate of the nozzle is calculated based on the corrected discharge coefficient, using the following formula:

[0124] ,

[0125] ,

[0126] in This represents the actual mass flow rate of the nozzle. Let be the geometric cross-sectional area of ​​the nozzle throat. This is the corrected outflow coefficient;

[0127] The results of all preceding signal processing and compensation steps (embodied in) This is integrated into the standard formula, thereby outputting a high-precision mass flow rate value that is based on classical theory and incorporates real-time state awareness and dynamic error compensation, achieving the final technical effect of the invention;

[0128] When the flow state is noncritical, a noncritical flow warning signal is issued, and the actual mass flow rate calculated in the previous stable critical flow state is output.

[0129] Mass flow calculation is still based on ISO and gas dynamics theory to ensure the physical correctness and standard consistency of the results. At the same time, by introducing dynamic correction coefficients, information such as the boundary layer state and throat condition changes sensed in real time is implicitly injected into the calculation process. Even if the nozzle experiences minor wear, deposition or environmental changes during use, the impact on flow measurement can still be automatically compensated through the preceding signal analysis steps. This promptly alerts the host system or operators that the current operating conditions do not meet the critical flow assumption, thus ensuring system safety and data reliability.

[0130] Example 2, refer to Figure 2 As a second embodiment of the present invention, a gas flow rate measurement error correction system based on a critical flow sonic nozzle includes:

[0131] The acquisition signal preprocessing module is used to acquire dynamic pressure, static pressure and static temperature multi-source data upstream of the sonic nozzle, and convert the voltage signal into a pressure pulsation sequence to complete noise reduction and standardization processing.

[0132] The boundary layer feature separation module is used to perform time-frequency analysis on the pressure pulsation signal through sliding window, wavelet transform and synchronous extrusion redistribution to obtain a sharpened average power spectrum. Based on cepstral analysis, it separates the turbulence source characteristics from the nozzle boundary layer system characteristics and extracts feature factors that reflect the throat boundary layer state.

[0133] The calibration and dynamic compensation module is used to combine the real-time Reynolds number with the pre-stored sensitivity coefficient to construct a boundary layer influence compensation model and calculate the corrected real-time outflow coefficient.

[0134] The critical flow discrimination and correction module is used to determine whether the nozzle is in a stable critical flow state, and outputs a high-precision mass flow rate based on the corrected discharge coefficient when the condition is met; otherwise, it issues an alarm.

[0135] This embodiment also provides a computer device applicable to the gas flow measurement error correction method based on a critical flow sonic nozzle, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the gas flow measurement error correction method based on a critical flow sonic nozzle as proposed in the above embodiment.

[0136] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0137] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the gas flow measurement error correction method based on a critical flow sonic nozzle as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0138] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for correcting gas flow measurement errors based on a critical flow sonic nozzle, characterized in that: Includes the following steps: Multi-source data is collected and preprocessed. The AC pulsation sequence is segmented and multiplied point by point with the Hamming window function to obtain the windowed signal sequence. The windowed signal sequence is then subjected to wavelet transform to determine the wavelet coefficients falling within the frequency axis neighborhood. The average energy along the time dimension is used to obtain the sharpened average power spectrum. Apply the natural logarithm to the sharpened average power spectrum and perform the inverse discrete Fourier transform to obtain the real cepstral sequence. Calculate the boundary layer characteristic factor, calculate the Reynolds number under the current flow condition, and convert it into the corrected outflow coefficient. The autocorrelation function of the real cepstrum in the high time delay interval is calculated, and the state of the critical flow sonic nozzle is judged. Error correction is performed based on the judgment results. The multi-source data includes voltage signals, static pressure, and static temperature data.

2. The gas flow rate measurement error correction method based on a critical flow sonic nozzle as described in claim 1, characterized in that: The process of segmenting the AC pulsating sequence and multiplying it point-by-point with a Hamming window function to obtain a windowed signal sequence, performing wavelet transform on the windowed signal sequence to determine the wavelet coefficients falling within the frequency axis neighborhood, and averaging the energy along the time dimension to obtain the sharpened average power spectrum includes: The AC pulsation sequence is segmented using the sliding window method to obtain a rectangular window sequence. The Hamming window function is then multiplied point by point with the rectangular window sequence to obtain the windowed signal sequence. Perform continuous wavelet transform on the windowed signal sequence to obtain wavelet coefficients, determine the instantaneous angular frequency by the partial derivative of the phase of the wavelet coefficients, and convert it into an instantaneous frequency; Define a frequency axis, aggregate all wavelet coefficients whose instantaneous frequencies fall within the neighborhood of the frequency axis, and obtain the time-frequency coefficients after synchronous squeezing transformation; The average energy of the time-frequency coefficients along the time dimension is calculated to obtain the sharpened average power spectrum.

3. The gas flow measurement error correction method based on a critical flow sonic nozzle as described in claim 2, characterized in that: The process of applying the natural logarithm to the sharpened average power spectrum and performing an inverse discrete Fourier transform to obtain a real cepstral sequence, and then calculating the boundary layer feature factors, includes: Apply the natural logarithm to each element of the sharpened average power spectrum to obtain the logarithmic power spectrum. Perform an inverse discrete Fourier transform on the logarithmic power spectrum to obtain the real cepstrum sequence. Set a time delay threshold, extract the low time delay components from the inverted frequency index from 1 to the time delay threshold in the real cepstrum, calculate the sum of squares, and take the square root to obtain the boundary layer feature factor.

4. The gas flow measurement error correction method based on a critical flow sonic nozzle as described in claim 3, characterized in that: The calculation of the Reynolds number under the current flow condition and its conversion into the corrected outflow coefficient includes: Calculate the Reynolds number under the current flow conditions based on static pressure and static temperature; Collect the historical Reynolds number of the clean nozzle and calculate the mean as the global feature reference value of the clean nozzle. Subtract the global feature reference value from the boundary layer feature factor to obtain the boundary layer feature deviation. The entire operating Reynolds number range of the nozzle is divided into continuous sub-intervals at equal intervals. The midpoint value of the sub-interval is calculated as the calibration Reynolds number. The compensation coefficient is calculated by combining the boundary layer characteristic deviation and the calibration Reynolds number. Based on the current Reynolds number, the baseline outflow coefficient is determined using the ISO 9300 standard, and multiplied by the current compensation coefficient to obtain the corrected outflow coefficient.

5. The gas flow measurement error correction method based on a critical flow sonic nozzle as described in claim 4, characterized in that: The calculation of the autocorrelation function of the real cepstrum in the high time delay interval, and the state determination of the critical flow sonic nozzle, includes: Preset a high time delay interval and calculate the autocorrelation function of the real cepstrum in the high time delay interval; Set a judgment threshold, and compare the maximum value of the autocorrelation function with the judgment threshold. If the maximum value of the autocorrelation function is greater than the judgment threshold, it is judged as a non-critical flow state; otherwise, it is judged as a stable critical flow state.

6. The gas flow measurement error correction method based on a critical flow sonic nozzle as described in claim 5, characterized in that: The error correction based on the judgment result includes: When the flow state is a steady critical flow state, the actual mass flow rate of the nozzle is calculated based on the corrected discharge coefficient. When the flow state is noncritical, a noncritical flow warning signal is issued, and the actual mass flow rate calculated in the previous stable critical flow state is output.

7. The gas flow measurement error correction method based on a critical flow sonic nozzle as described in claim 1, characterized in that: The process of collecting and preprocessing multi-source data includes: Upstream of the sonic nozzle, smart sensors are used to collect multi-source data and perform preprocessing. The intelligent sensors include dynamic pressure, absolute pressure, and temperature sensors; The preprocessing includes converting the voltage signal value into the corresponding instantaneous absolute pressure, splicing them into a pressure sequence in chronological order, applying a first-order recursive digital high-pass filter to the pressure sequence to obtain an AC pulsation sequence. The static pressure and static temperature are denoised and standardized.

8. A gas flow measurement error correction system based on a critical flow sonic nozzle, used to implement the gas flow measurement error correction method based on a critical flow sonic nozzle as described in any one of claims 1 to 7, characterized in that: include: The acquisition signal preprocessing module is used to acquire dynamic pressure, static pressure and static temperature multi-source data upstream of the sonic nozzle, and convert the voltage signal into a pressure pulsation sequence to complete noise reduction and standardization processing. The boundary layer feature separation module is used to perform time-frequency analysis on the pressure pulsation signal through sliding window, wavelet transform and synchronous extrusion redistribution to obtain a sharpened average power spectrum. Based on cepstral analysis, it separates the turbulence source characteristics from the nozzle boundary layer system characteristics and extracts feature factors that reflect the throat boundary layer state. The calibration and dynamic compensation module is used to combine the real-time Reynolds number with the pre-stored sensitivity coefficient to construct a boundary layer influence compensation model and calculate the corrected real-time outflow coefficient. The critical flow discrimination and correction module is used to determine whether the nozzle is in a stable critical flow state, and outputs a high-precision mass flow rate based on the corrected discharge coefficient when the condition is met; otherwise, it issues an alarm.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the gas flow measurement error correction method based on a critical flow sonic nozzle as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the gas flow measurement error correction method based on a critical flow sonic nozzle as described in any one of claims 1 to 7.

Images