Security type intelligent diaphragm gas meter

Abstract

The application discloses a security and protection type intelligent diaphragm gas meter, and particularly relates to the technical field of gas meter safety protection, which is used to solve the problem that the existing gas meter cannot detect the combustible mixed gas caused by air backflow due to negative pressure when the valve is quickly closed; the flow monitoring module is used to monitor the flow in real time and trigger the valve closing; the pressure gradient module is used to collect pressure values at multiple positions to generate a pressure gradient vector during the valve closing; the spectrum shift module is used to analyze the pressure values by Hilbert-Huang transform to extract the marginal spectrum energy barycentric offset; the air backflow risk is determined by the air backflow judging module based on the included angle between the pressure gradient direction angle and the maximum strength direction of the leakage signal and the spectrum offset; the leakage analyzing module is used to calculate the local curvature distribution of the pressure wave propagation wave front through the pressure oscillation waveform, and multiple discrete curvature maximum points are identified to determine the multiple leakage point cooperative backflow; the joint control module is used to switch the two-stage valve closing mode and trigger the linkage ventilation instruction according to the risk determination result, so that the combustible mixed gas hidden danger in the pipeline is eliminated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gas meter safety protection, and particularly to an intelligent security film type gas meter. BACKGROUND

[0002] The existing security type gas meter generally configures an automatic valve closing function. When detecting gas leakage or flow anomaly, the risk is controlled by quickly cutting off the gas source. The valve response mechanism is usually designed according to international standards, focusing on optimizing the valve closing speed and sealing performance to ensure that the gas output is blocked in the shortest time. In the industry technical cognition, the valve closing action is regarded as the terminal measure of safety protection, and the design criteria are focused on improving the closing efficiency and reliability.

[0003] However, the existing technology ignores the fluid dynamics effect caused by the valve closing operation: in the process of quickly cutting off the gas source, the sudden stop of the gas flow in the downstream pipeline will cause a sudden change in local pressure, which may form a significant negative pressure in the pipeline, causing external air to suck into the closed pipe network through the leakage point. The mixture of air and residual gas will form a combustible gas, and the mixture is distributed inside the pipeline. The leakage sensor installed at the end of the traditional table cannot directly detect such hidden risks. There is no effective response mechanism for new safety hazards induced by fluid transient process after valve closing, forming a breakpoint in the protection chain. SUMMARY

[0004] The present application provides a security type intelligent film type gas meter to solve the technical problems in the prior art.

[0005] The technical solution of the present application to solve the above technical problems is as follows:

[0006] The present application provides the following technical solutions:

[0007] A security type intelligent film type gas meter, comprising:

[0008] A flow monitoring module for real-time monitoring of gas flow value, generating a valve closing action signal when detecting that the gas flow value exceeds a preset leakage threshold;

[0009] A pressure gradient module for starting valve closing operation in response to the valve closing action signal, and generating a pressure gradient vector according to the spatial distribution of pressure values collected at different positions of the pipeline during the valve closing operation; synchronously acquiring a leakage detection signal;

[0010] A spectrum shift module for analyzing the pressure values by Hilbert-Huang transform and extracting a marginal spectrum energy barycenter shift amount;

[0011] A suction judgment module for judging that there is a risk of air suction when the included angle between the direction angle of the pressure gradient vector and the maximum intensity direction of the leakage detection signal is less than a critical angle and the marginal spectrum energy barycenter shift amount is greater than a frequency shift threshold.

[0012] a leakage analysis module, configured to calculate a local curvature distribution of a pressure wave propagation wave front by an oscillation waveform of the pressure value, and determine that multiple leakage points are in coordination with backflow when there are at least two discrete maximum points in the local curvature distribution;

[0013] a joint control module, configured to switch the valve closing action to a two-stage mode and trigger a joint ventilation instruction based on the determination results of the air backflow risk and the multiple leakage points in coordination with backflow.

[0014] Further, the gas flow value is monitored in real time, and a valve closing action signal is generated when it is detected that the gas flow value exceeds a preset leakage threshold, including:

[0015] The gas flow value is obtained in real time by a diaphragm metering mechanism;

[0016] The gas flow value is compared with the preset leakage threshold;

[0017] When the gas flow value exceeds the preset leakage threshold, a valve closing action signal is generated.

[0018] Further, after the state that the gas flow value exceeds the preset leakage threshold lasts for a preset time length, the operation of generating the valve closing action signal is performed.

[0019] Further, a valve closing operation is started in response to the valve closing action signal, and during the valve closing operation, pressure values at different positions of the pipeline are collected and a pressure gradient vector is generated according to the spatial distribution of the pressure values; a leakage detection signal is synchronously obtained, including:

[0020] A valve closing action is started in response to the valve closing action signal;

[0021] During the continuous valve closing action, pressure values at different positions of the pipeline are collected by pressure sensors arranged at the head end, the middle end and the tail end of the pipeline;

[0022] According to the spatial distribution of the pressure values at the head end, the middle end and the tail end of the pipeline, a pressure gradient vector is calculated;

[0023] Synchronously, a leakage detection signal is obtained by a distributed gas sensor array.

[0024] Further, the pressure values are analyzed by Hilbert-Huang transform to extract a marginal spectrum energy barycenter offset, including:

[0025] The pressure values are subjected to empirical mode decomposition to generate a plurality of intrinsic mode function components;

[0026] The Hilbert transform is performed on each intrinsic mode function component to obtain an instantaneous frequency and an instantaneous amplitude;

[0027] A Hilbert marginal spectrum is constructed based on the instantaneous frequency and the instantaneous amplitude;

[0028] calculating an energy centroid frequency of the Hilbert marginal spectrum;

[0029] comparing the energy centroid frequency with a reference centroid frequency to obtain a marginal spectrum energy centroid shift.

[0030] Further, performing a Hilbert transform on each of the proper modal function components to obtain an instantaneous frequency and an instantaneous amplitude, comprising:

[0031] constructing an analytic signal for the proper modal function component;

[0032] extracting a phase angle function of the analytic signal, calculating a first derivative of the phase angle function with respect to time, and dividing the first derivative of the phase angle function with respect to time by twice the value of pi to obtain the instantaneous frequency;

[0033] calculating a sum of squares of a real part and an imaginary part of the analytic signal, and performing an arithmetic square root operation on the sum of squares to obtain the instantaneous amplitude.

[0034] Further, when an included angle between a directional angle of the pressure gradient vector and a maximum intensity azimuth of the leak detection signal is less than a critical angle and the marginal spectrum energy centroid shift is greater than a frequency shift threshold, determining that the air reverse suction risk occurs, comprising:

[0035] calculating the directional angle of the pressure gradient vector;

[0036] determining the maximum intensity azimuth of the leak detection signal;

[0037] calculating the included angle between the directional angle and the maximum intensity azimuth;

[0038] obtaining the marginal spectrum energy centroid shift;

[0039] comparing the included angle with the critical angle and comparing the marginal spectrum energy centroid shift with the frequency shift threshold;

[0040] when the included angle is less than the critical angle and the marginal spectrum energy centroid shift is greater than the frequency shift threshold, generating an air reverse suction risk determination result.

[0041] Further, calculating a local curvature distribution of a pressure wave propagation wave front through an oscillation waveform of pressure values, and when there are at least two discrete maximum value points in the local curvature distribution, determining that multiple leak points cooperatively suck back, comprising:

[0042] obtaining an oscillation waveform of pressure values at different positions of the pipeline;

[0043] reconstructing a propagation process of the pressure wave in the pipeline based on the oscillation waveform;

[0044] identifying a morphological change feature of the pressure wave propagation wave front;

[0045] Quantify the local bending degree of the wave front according to the morphological change characteristics;

[0046] Scan the local bending degree distribution in the pipe space range;

[0047] Mark the position points where the local bending degree is prominent;

[0048] Analyze the spatial distribution relationship of the marked position points;

[0049] When there are at least two mutually isolated marked position points, a plurality of leakage point collaborative suction determination results are generated.

[0050] Further, based on the air suction risk and the determination result of the plurality of leakage point collaborative suction, the valve closing action is switched to a two-stage mode and a linkage ventilation instruction is triggered, including:

[0051] Receive the air suction risk determination result and the plurality of leakage point collaborative suction determination result;

[0052] When there is an air suction risk, a linkage ventilation instruction is generated, which is used to dilute the residual gas in the pipeline;

[0053] When there is an air suction risk and a plurality of leakage point collaborative suction at the same time, the valve closing action is switched to a two-stage mode.

[0054] Further, the two-stage mode is specifically: in the first stage, the valve is closed to a first opening degree at a first speed, and the first opening degree maintains the gas flow value below the lower limit of the mixed gas explosion; in the second stage, the valve is closed to a full closed state at a second speed lower than the first speed.

[0055] The beneficial effects of the present application are:

[0056] 1. By dynamically sensing the fluid transient effect in the closing valve process, an air suction risk identification system is constructed, which is different from the traditional gas meter which only relies on the monitoring logic of the end flow and the leakage sensor. In the closing valve operation, the pressure values at multiple positions in the pipeline are collected synchronously, the pressure gradient vector is generated to represent the negative pressure propagation direction, the marginal spectrum energy barycenter offset extracted by Hilbert-Huang transform is combined to accurately capture the gas composition change characteristics; At the same time, the local curvature distribution analysis of the pressure wave propagation wave front is used to identify the wave front distortion effect caused by multiple discrete leakage points, realize the real-time determination of the air suction risk induced by the closing valve and the collaborative effect of multiple leakage points, and solve the problem of hidden combustible gas caused by external air suction due to rapid closing valve.

[0057] 2. The valve closing operation is upgraded from terminal safety measures to a dynamic risk control process. Based on the air suction risk judgment result, the control module intelligently switches the two-stage valve closing mode: in the first stage, the speed is reduced to close to maintain the gas flow below the lower limit of explosion, and the negative pressure surge path is blocked; in the second stage, the ventilation instruction is triggered synchronously during the low-speed closing process, and the residual gas in the pipeline is actively diluted, thereby inhibiting the generation conditions of flammable gas mixture from the source, converting the "passive shutdown" of the traditional gas meter into "active hazard elimination", and significantly improving the intrinsic safety level of the closed pipe network system. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 FIG. 1 is a structural schematic diagram of a security and protection type intelligent diaphragm gas meter according to the present application;

[0059] Figure 2 FIG. 4 is a flowchart of a plurality of leakage point collaborative suction determination according to the present application. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0061] Embodiment: Figure 1 FIG. 1 is a structural schematic diagram of a security and protection type intelligent diaphragm gas meter according to the present application. The security and protection type intelligent diaphragm gas meter comprises:

[0062] A flow monitoring module is configured to monitor the gas flow value in real time and generate a valve closing action signal when detecting that the gas flow value exceeds a preset leakage threshold;

[0063] A pressure gradient module is configured to start the valve closing operation in response to the valve closing action signal, acquire pressure values at different positions of the pipeline during the valve closing operation, and generate a pressure gradient vector according to the spatial distribution of the pressure values; and simultaneously acquire a leakage detection signal;

[0064] A spectrum shift module is configured to analyze the pressure values by Hilbert-Huang transform and extract a marginal spectrum energy barycenter shift amount;

[0065] A suction judgment module is configured to judge that an air suction risk occurs when the included angle between the direction angle of the pressure gradient vector and the maximum intensity direction of the leakage detection signal is less than a critical angle and the marginal spectrum energy barycenter shift amount is greater than a frequency shift threshold;

[0066] The leakage analysis module is configured to calculate a local curvature distribution of a pressure wave propagation wave front through an oscillation waveform of the pressure value, and determine that multiple leakage points are in coordination with backflow when there are at least two discrete maximum points in the local curvature distribution.

[0067] The joint control module is configured to switch the valve closing action to a two-stage mode and trigger a joint ventilation instruction based on the determination results of the air backflow risk and the multiple leakage points in coordination with backflow.

[0068] The gas flow value is monitored in real time, and a valve closing action signal is generated when it is detected that the gas flow value exceeds a preset leakage threshold, and the valve closing action signal is specifically implemented as follows:

[0069] The diaphragm metering mechanism is composed of a metering cavity, a flexible diaphragm, and a crank connecting rod assembly. The metering cavity is connected with a user-side gas equipment through a gas inlet pipeline, and the flexible diaphragm separates the metering cavity into two gas chambers with periodic volume changes. When the gas flows through the metering cavity, the flexible diaphragm reciprocally deforms under the action of the pressure difference, and drives the crank connecting rod assembly to convert the diaphragm displacement into rotary motion. The rotary motion is transmitted to the flow metering disc through the transmission gear set, and the flow metering disc corresponds to a fixed volume of gas passing through every fixed angle of rotation. An optical encoder is installed at the rotation shaft of the flow metering disc, and the optical encoder outputs a pulse signal sequence in real time by detecting the number of light transmission grids on the disc. The control unit receives the pulse signal sequence, and calculates the current gas flow value by arithmetic multiplication according to the number of pulses and the calibration coefficient of the gas volume corresponding to a unit pulse. The unit of the gas flow value is cubic meters per hour.

[0070] The preset leakage threshold is set by the following method: first, the basic flow fluctuation range of the gas meter in the no-leakage state is obtained, the basic flow fluctuation range is obtained by continuously monitoring the typical gas use behavior of the user for thirty days, and the typical gas use behavior includes the scenarios of igniting the cooking utensil and starting and stopping the water heater; the maximum value of the basic flow fluctuation range is taken twice as the leakage determination reference value; the leakage determination reference value is multiplied by a safety margin coefficient to obtain the preset leakage threshold, and the safety margin coefficient is in the range of 1.2 to 1.5, which is determined according to the statistical characteristics of the pressure fluctuation of the gas pipeline. For example, for a gas meter with a range of 6 cubic meters per hour, the maximum value of the basic flow fluctuation range is 0.02 cubic meters per hour, the leakage determination reference value is 0.04 cubic meters per hour, and the safety margin coefficient is 1.3, then the preset leakage threshold is 0.052 cubic meters per hour.

[0071] The control unit continuously compares the real-time acquired gas flow value with the preset leakage threshold. The comparison operation is realized by a digital comparator circuit, the first input end of the digital comparator circuit is connected to the output end of the analog-to-digital converter, and the analog-to-digital converter converts the gas flow value into a digital signal; the second input end of the digital comparator circuit is connected to the output end of the memory, and the memory stores the digital code of the preset leakage threshold; the output end of the digital comparator circuit outputs a high level to indicate that the gas flow value exceeds the preset leakage threshold.

[0072] When the digital comparator circuit outputs a high level, the timer triggers to start the cumulative timing. The timer uses a crystal oscillator as the clock source, and the timing accuracy is 0.1 seconds. The preset time length is determined by the following method: analyzing the time required for the residual gas in the gas pipeline to be completely discharged, and the time required for the residual gas to be completely discharged is calculated based on the quotient of the pipeline volume and the flow rate; take 0.8 times of the time required for the residual gas to be completely discharged as the preset time length reference value; adjust the preset time length reference value according to the pipeline length, and the adjustment coefficient is the pipeline length divided by 10 meters (pipeline length ≥ 1 meter). For example, for a pipeline with a length of 5 meters, the residual gas needs 20 seconds to be discharged, and the preset time length reference value is 16 seconds, and the adjustment coefficient is 0.5, and the final preset time length is 8 seconds.

[0073] When the cumulative time of the timer reaches the preset time length, the control unit generates a valve closing action signal. The valve closing action signal is a 12-volt direct current pulse voltage signal with a pulse width of 100 milliseconds, which is converted and generated by the driving circuit from a 5-volt logic level. The enable end of the driving circuit is connected to the output end of the timer, and when the enable end receives a high level signal and the continuous time reaches the standard, the driving circuit outputs a 12-volt pulse.

[0074] The continuous time determination mechanism is used to exclude transient flow interference. During the cumulative process of the timer, if the gas flow value falls below the preset leakage threshold and lasts for 0.5 seconds, the timer is immediately reset. The transient flow interference scenarios include: flow peak caused by initial gas pressure fluctuation of the stove (duration < 2 seconds), and pipeline pressure transient caused by water hammer effect (duration < 0.3 seconds).

[0075] In response to the valve closing action signal, the valve closing operation is started, and during the valve closing operation, the pressure values at different positions of the pipeline are collected, and a pressure gradient vector is generated according to the spatial distribution of the pressure values; the leakage detection signal is acquired synchronously, which is specifically implemented as:

[0076] The control unit sends a start command to the valve drive mechanism upon receiving the valve closing signal. The valve drive mechanism consists of a stepper motor, a gear set and a valve core position sensor. The stepper motor converts the rotational motion into linear displacement of the valve stem through the gear set to push the valve core to the closing direction upon receiving the start command. The valve core position sensor detects the displacement of the valve core in real time and feeds back to the control unit to form a closed-loop control. The valve closing action duration is defined as the entire time period from the start of the valve core movement to the full-closed position. The time period varies according to the type of valve, for example, the full stroke time of a 20mm diameter ball valve is usually 3-5 seconds.

[0077] The pressure values at the pipe head, middle and end are synchronously collected by pressure sensors installed at the pipe head, middle and end during the valve closing action duration. The pipe head refers to the pipe section closest to the valve within a range of 0.5 meters from the valve outlet; the pipe middle refers to the pipe section within a range of ±10% of the midpoint of the total length of the pipe; and the pipe end refers to the pipe section farthest from the valve within a range of 1 meter from the user terminal equipment. The pressure sensor uses a piezoresistive sensing element with a range of -10 kPa to +50 kPa and a sampling frequency of 100 Hz. The three sensors are time-synchronized through a unified clock source with a synchronization accuracy of 0.1 ms. The control unit reads and stores the pressure value data at a cycle of 10 ms during the collection process, and the pressure value unit is kPa.

[0078] To calculate the pressure gradient vector according to the spatial distribution of the pressure values at the pipe head, middle and end, a pipe spatial coordinate system is first established: the pipe head is taken as the coordinate origin, and the pipe axial direction is taken as the positive direction of the X axis. The coordinates of the pipe head, middle and end are (0, 0, 0), (L / 2, 0, 0) and (L, 0, 0) respectively, where L is the total length of the pipe. The calculation process includes the following steps:

[0079] The difference between the middle pressure value and the head pressure value is calculated as the first pressure change; the difference between the end pressure value and the middle pressure value is calculated as the second pressure change; the second pressure change is multiplied by the value 2; the multiplication result is divided by the total length L of the pipe to obtain the component value of the pressure gradient vector in the pipe direction; the remaining direction components of the pressure gradient vector are set to zero.

[0080] For example, when the pipe length L is equal to 10 meters and the end pressure value is 1 kPa higher than the middle pressure value, the component value of the pressure gradient vector in the pipe direction is 0.2 kPa per meter.

[0081] Synchronous acquisition of the leakage detection signal is achieved by a distributed gas sensor array. The gas sensor array is composed of multiple gas concentration detection units. The detection units are distributed at equal intervals along the pipeline direction. The interval is determined according to the length of the pipeline. Usually, one detection unit is arranged every 2 meters. Each detection unit contains a methane sensor and a signal conditioning circuit. The methane sensor adopts the catalytic combustion principle to detect a range of 0 to 100% LEL. The signal conditioning circuit amplifies the millivolt-level voltage signal output by the sensor and converts it into a 4-20 mA standard current signal. The control unit acquires the signals of each detection unit through a multiplexer in a round-robin manner. The round-robin acquisition period is synchronized with the pressure value acquisition period and is set to 10 milliseconds. The leakage detection signal is defined as the spatial orientation corresponding to the maximum amplitude of the output signal of each detection unit. The orientation is determined by the position number of the detection unit. For example, when the output signal amplitude of the detection unit numbered 3 is the largest and the unit is located 6 meters away from the head end, the maximum intensity orientation of the leakage detection signal is recorded as 6 meters along the pipeline axis.

[0082] The pressure gradient vector calculation process sets up data validity verification: when any value of the pressure values at three positions exceeds the range, for example, is lower than -10 kPa or higher than +50 kPa, the data acquisition is discarded and re-acquired; when the absolute value of the pressure difference between the head end and the tail end is less than 0.5 kPa, it is determined as an invalid gradient and a zero vector is output. When all detection unit outputs are below the detection threshold, for example, 2% LEL, it is determined that there is no leakage signal and an empty orientation marker is output.

[0083] The spatial position calibration is achieved by pipeline mapping data. The total length L of the pipeline is obtained by actual measurement during installation, with a measurement accuracy of centimeter level. The coordinates of each sensor position are recorded during initialization of the control system. For example, the head end sensor is recorded as (0, 0, 0), the tail end sensor is recorded as (5.0, 0, 0), and the end sensor is recorded as (10.0, 0, 0). The gas detection unit positions are recorded according to the actual installation distance. The number database number and spatial position form a mapping relationship.

[0084] The pressure values are analyzed by Hilbert-Huang transform, and the marginal spectrum energy barycenter offset is extracted. The specific implementation is as follows:

[0085] When the pressure values collected by the pressure sensor are decomposed by empirical mode decomposition, the adaptive iterative filtering algorithm is used to process the discrete time series data. The pressure value data is stored in a ring buffer with a sampling interval of 10 milliseconds. Each time, 1024 consecutive sampling points are taken to form an analysis window. The empirical mode decomposition process includes the following steps: first, identify the local maximum points and local minimum points in the pressure value time series. The local maximum point is defined as a data point greater than the adjacent two sampling points, and the local minimum point is defined as a data point less than the adjacent two sampling points. Connect all local maximum points by cubic spline interpolation to form the upper envelope line, and connect all local minimum points to form the lower envelope line. Calculate the arithmetic mean of the upper envelope line and the lower envelope line at each sampling point to obtain the mean envelope line. Subtract the mean envelope line from the original pressure value sequence to obtain the intermediate component. Repeat the above process until the intermediate component meets the two judgment conditions of intrinsic mode function: condition one is that the number of zero-crossing points and the number of extreme points in the entire data range differ by no more than one, and condition two is that the difference between the mean of the upper envelope line and the lower envelope line and zero at any time is less than 0.01 kilopascal. When the conditions are met, output the intrinsic mode function component. Subtract the component from the original sequence, and continue to decompose the remaining sequence until the remaining sequence is a monotonic function. For example, after decomposing a certain segment of pressure fluctuation data containing 1024 sampling points, 3 intrinsic mode function components and 1 residual component may be obtained. The number of iterations in the decomposition process is usually 5 to 10 times.

[0086] When performing Hilbert transform on each intrinsic mode function component, the following operations are implemented in the digital signal processor: read the discrete data point sequence of the intrinsic mode function component, and perform fast Fourier transform on the sequence to obtain the frequency domain representation; set the negative frequency part of the frequency domain representation to zero, multiply the positive frequency part by a coefficient of 2, and keep the zero frequency component unchanged; perform inverse Fourier transform on the processed frequency domain data to obtain a complex signal sequence, which is the analytic signal. The real part of the analytic signal is equal to the original intrinsic mode function component, and the imaginary part is the Hilbert transform result thereof. This process needs to meet the sampling theorem requirement, and a sampling frequency of 100 Hz corresponds to a highest analysis frequency of 50 Hz.

[0087] When extracting the instantaneous frequency from the analytic signal, the following calculation steps are performed: read the real part sequence and the imaginary part sequence of the analytic signal; calculate the ratio of the imaginary part value to the real part value for each sampling point; perform four-quadrant arctangent operation on the ratio to obtain the phase angle function value, which ranges from negative pi to positive pi; calculate the difference between the phase angle function values of adjacent two sampling points, where adjacent sampling points refer to consecutive sampling points in time sequence; divide the difference by the sampling time interval of 0.01 seconds to obtain the phase angle change rate; divide the phase angle change rate by 2 and multiply it by pi to obtain the instantaneous frequency value, which is in units of hertz. For example, when the phase angle difference of adjacent sampling points is 0.314 radians, the phase angle change rate is 31.4 radians per second, and the instantaneous frequency is 5 Hz.

[0088] When extracting instantaneous amplitude from the parsed signal, the following calculation steps are performed: reading the real part sequence and the imaginary part sequence of the parsed signal; squaring the real part value of each sampling point; squaring the imaginary part value; adding the real part square value and the imaginary part square value to obtain a square sum; performing an arithmetic square root operation on the square sum result to obtain the instantaneous amplitude, and the unit of the instantaneous amplitude is kilopascal. For example, when the real part of a certain sampling point is 0.8 kilopascal and the imaginary part is 0.6 kilopascal, the real part square is 0.64, the imaginary part square is 0.36, the square sum is 1.0, and the arithmetic square root is 1.0 kilopascal.

[0089] When constructing the Hilbert marginal spectrum based on the instantaneous frequency and the instantaneous amplitude of all sampling points, the following process is performed: setting the frequency analysis range to 0 hertz to 50 hertz, and the upper limit of the frequency is determined according to the Nyquist sampling theorem; dividing the frequency range into 500 equally spaced frequency bands, and the width of each frequency band is 0.1 hertz; for each frequency band, counting the sampling points whose instantaneous frequency falls within the frequency band, and the instantaneous frequency falling condition is defined as the instantaneous frequency value being greater than or equal to the lower limit of the frequency band and less than the upper limit of the frequency band; summing the instantaneous amplitudes corresponding to these sampling points to obtain the cumulative amplitude of the frequency band; calculating the sum of the cumulative amplitudes of all frequency bands; dividing the cumulative amplitude of each frequency band by the sum to obtain the normalized cumulative amplitude; and the distribution of the normalized cumulative amplitudes of all frequency bands constitutes the Hilbert marginal spectrum. For example, if there are 20 sampling points in the 10.0 hertz to 10.1 hertz frequency band, the sum of their instantaneous amplitudes is 15.0 kilopascal, and the total cumulative amplitude is 3000 kilopascal, then the normalized cumulative amplitude of the frequency band is 0.005.

[0090] When calculating the energy center frequency of the Hilbert marginal spectrum, a weighted average algorithm is used: taking the center frequency value of each frequency band, which is equal to the lower limit of the frequency band plus 0.05 hertz; multiplying the center frequency value by the normalized cumulative amplitude of the corresponding frequency band to obtain a weighted value; summing the weighted values of all frequency bands to obtain a weighted sum; and the weighted sum is the energy center frequency, with the unit of hertz. For example, when the main energy is concentrated near 10 hertz, if the center frequency of 10.05 hertz corresponds to an amplitude of 0.01 and the center frequency of 10.15 hertz corresponds to an amplitude of 0.008, then the weighted sum is 10.05 x 0.01 + 10.15 x 0.008 +... = 10.5 hertz.

[0091] The reference center frequency is determined by the following method: continuously collecting 30 groups of pressure value samples under normal operating conditions of the gas system, and each group of samples contains 1024 sampling points; performing the above Hilbert marginal spectrum analysis process on each group of samples and calculating the energy center frequency; taking the arithmetic mean of the 30 energy center frequency values as the reference center frequency, and the arithmetic mean calculation process is to divide the sum by 30. The reference center frequency is stored in the non-volatile memory, and the update period is three months, which is set according to the seasonal variation characteristics of the gas system.

[0092] The marginal spectral energy centroid shift is calculated as the algebraic difference between the current energy centroid frequency and the reference centroid frequency. The algebraic difference is equal to the current energy centroid frequency minus the reference centroid frequency, and the shift is in units of hertz. A positive shift indicates that the energy distribution has moved in the high frequency direction, and a negative shift indicates that the energy distribution has moved in the low frequency direction. For example, if the reference centroid frequency is 25.0 hertz and the current energy centroid frequency is 28.0 hertz, then the shift is +3.0 hertz.

[0093] Data validity checks are implemented during the process: when the instantaneous frequency calculation yields a negative value, the data point is discarded, as negative frequencies have no physical meaning; when the instantaneous amplitude exceeds the sensor range by plus 50 kilopascals or falls below the sensor range by minus 10 kilopascals, the data is marked as invalid; when the energy centroid frequency exceeds the range of 0 to 50 hertz, the empirical mode decomposition is re-performed. The invalid data is handled by discarding the abnormal sample point and replacing it with the value of the previous valid sample point. All calculations are performed in a digital signal processor, and the processing delay from data input to output of the shift is measured to be less than 100 milliseconds.

[0094] When the included angle between the direction angle of the pressure gradient vector and the azimuth of the maximum strength of the leak detection signal is less than the critical angle and the marginal spectral energy centroid shift is greater than the frequency shift threshold, it is determined that there is a risk of air suck-back, and the specific implementation is as follows:

[0095] When calculating the direction angle of the pressure gradient vector, the operation is performed in the pipe spatial coordinate system. The pressure gradient vector is a three-dimensional vector, and its direction angle is defined as the included angle between the vector and the positive direction of the pipe axis. Since the pressure gradient vector only has a component in the direction of the pipe, the direction angle is calculated as follows: when the component value of the pressure gradient vector in the direction of the pipe is positive, the direction angle is 0 degrees, and when the component value is negative, the direction angle is 180 degrees. The calculation process is performed in a microcontroller, for example, when the component value of the pressure gradient vector in the direction of the pipe is 0.2 kilopascals per meter, the direction angle is 0 degrees, and when the component value is negative 0.2 kilopascals per meter, the direction angle is 180 degrees. The angle value is stored as a single-precision floating-point number, with one decimal place of precision.

[0096] The output signal of the distributed gas sensor array is processed to determine the maximum intensity direction of the leak detection signal. The maximum intensity direction is defined as the spatial position corresponding to the detection unit with the maximum amplitude of the output signal, which is obtained by the following steps: the current signal of each detection unit is collected, the current signal ranges from 4 to 20 mA; the current signal is converted into a digital amplitude value; the amplitudes of all detection units are compared; the detection unit number corresponding to the maximum amplitude is recorded; and the spatial coordinates are determined according to the pre-stored mapping relationship between the number and the position. For example, when the detection unit with the number 3 has the maximum amplitude and its position coordinates are 6 meters in the axial direction of the pipeline, the maximum intensity direction is recorded as 6 meters. When the amplitudes of multiple detection units are the same, the detection unit with the maximum position coordinate value is selected as the maximum intensity direction.

[0097] When calculating the included angle between the direction angle and the maximum intensity direction, the spatial position needs to be converted into an angular quantity. The azimuth angle conversion method is: the axial distance value corresponding to the maximum intensity direction is divided by the total length L of the pipeline, and the division result is multiplied by one hundred and eighty to obtain the angle value. For example, when the length of the pipeline is 10 meters and the position of the leakage point is 6 meters, the azimuth angle is 108 degrees. The included angle is calculated by taking the absolute value of the difference between the direction angle and the azimuth angle. For example, when the direction angle is 0 degrees and the azimuth angle is 108 degrees, the included angle is 108 degrees. The calculation result is stored as an unsigned integer, with the unit of degrees.

[0098] When obtaining the center of gravity offset of the marginal spectrum energy, the output value of the spectrum shift module is read through the data bus. The offset unit is hertz, represented by a signed single-precision floating-point number, and the numerical range is limited between -50.0 hertz and 50.0 hertz. The reading period is set to 10 milliseconds synchronously with the pressure collection period, and the data is transmitted through the SPI interface.

[0099] The critical angle setting method is as follows: the angle deviation data between the pressure gradient direction and the azimuth of the leakage point under different leakage conditions are obtained through fluid dynamics simulation, the minimum angle deviation value is multiplied by a safety factor of 0.8 to obtain the reference critical angle, and then the pipeline material is adjusted: the reference value is used for metal pipelines, and 5 degrees are added for non-metal pipelines. For example, when the simulation minimum angle is 15 degrees, the reference critical angle is 12 degrees, and the plastic pipeline uses 17 degrees. The critical angle is stored in the EEPROM, and the update period is six months.

[0100] The frequency shift threshold determination method is as follows: the center of gravity offset of the marginal spectrum energy under normal working conditions for thirty consecutive days is recorded, the standard deviation of these data is calculated, the standard deviation is multiplied by a coefficient of three as the basic threshold, and a safety margin of 10% is added. For example, when the standard deviation is 0.5 hertz, the basic threshold is 1.5 hertz, and the final frequency shift threshold is 1.65 hertz. The threshold is automatically updated once a month.

[0101] The comparison operation is realized by a hardware comparison circuit: a first comparator receives the included angle value and a critical angle, and outputs a high level when the included angle is less than the critical angle; a second comparator receives the offset and a frequency shift threshold, and outputs a high level when the offset is greater than the frequency shift threshold; and the outputs of the two comparators are connected to an AND gate circuit. For example, the critical angle is set to 12 degrees, the first comparator outputs a high level when the included angle is 10 degrees; the frequency shift threshold is set to 1.65 Hz, the second comparator outputs a high level when the offset is 2.0 Hz, and the AND gate outputs a high level.

[0102] When generating the air backflow risk determination result, the AND gate output signal is converted into a data packet. When the AND gate outputs a high level, a risk exists identifier 0xAA is generated, and when the AND gate outputs a low level, a no risk identifier 0x55 is generated. The data packet contains a time stamp and a check code, and is transmitted to the joint control module at a rate of 500 kbps through the CAN bus. For example, the generated data packet format is [0xAA, time stamp, CRC8].

[0103] An abnormality processing mechanism is set: when the maximum intensity direction is an empty direction marker, that is, all detection unit signals are lower than 2% LEL, a no risk determination is directly output; when the offset exceeds the effective range, an arithmetic mean of the last three effective offsets is used to replace it; and when the pipeline length is less than 1 meter, the included angle is fixed to 0 degrees. The abnormal state is recorded in the system log.

[0104] A three-level verification mechanism of the determination result: after the first determination, a second determination is performed after a delay of 100 milliseconds, and when the two results are consistent, the final result is confirmed; when the two results are inconsistent, a third determination is performed, and the same result of the two times is taken as the final output. For example, when the first determination has a risk and the second determination has no risk, the third determination is started to take the same result. The maximum time consumption of the verification process is 300 milliseconds.

[0105] Dynamic adjustment of the critical angle and the frequency shift threshold: the false alarm rate and the missed alarm rate are counted every twenty-four hours, the false alarm rate is defined as the ratio of misjudging a no risk event as a risk event, and the missed alarm rate is defined as the ratio of missing a risk event. When the false alarm rate exceeds 5%, the critical angle is increased by 1 degree; and when the missed alarm rate exceeds 2%, the frequency shift threshold is increased by 0.1 Hz. The adjustment amplitude does not exceed ±20% of the initial value. For example, the initial critical angle is 12 degrees, and when the false alarm rate is 6%, the critical angle is adjusted to 13 degrees.

[0106] The physical meaning of the included angle between the pressure gradient vector direction angle and the leakage direction: when the included angle is less than the critical angle, it indicates that the leakage point is located in the direction of the negative pressure wave caused by the valve closing; when the offset is greater than the frequency shift threshold, it indicates that the gas composition in the pipeline changes, causing the sound speed to change, and the simultaneous occurrence of the two indicates that air is being sucked back into the pipeline.

[0107] Figure 2The flow chart of the multiple leak point cooperative backflow determination of the application is given, the local curvature distribution of the pressure wave propagation wave front is calculated through the oscillation waveform of the pressure value, when there are at least two discrete maximum points in the local curvature distribution, it is determined that multiple leak point cooperative backflow occurs, and the specific implementation is:

[0108] When the oscillation waveform of the pressure value at different positions of the pipeline is obtained, the historical data buffer of the head-end, middle-end and tail-end pressure sensors is called. The oscillation waveform refers to the dynamic pressure change curve recorded during the valve closing process, the time span is 100 milliseconds before the valve starts to close to 500 milliseconds after it is completely closed, and the sampling frequency is 100 Hz. 600 sampling point data are saved at each position, and the data format is single-precision floating point number with timestamp, unit kilopascal. The waveform data is transmitted to the signal processor through the direct memory access channel, and the transmission process does not occupy the central processing unit resources.

[0109] When the pressure wave propagation process is reconstructed based on the oscillation waveform, a time-space interpolation algorithm is used. First, the pipeline space model is established: taking the head-end of the pipeline as the coordinate zero point and the tail-end as the terminal point, the total length L is divided into 100 equal grid points. For each time point (interval 10 milliseconds), the following operations are performed: reading the pressure values at three positions; calculating the pressure values of 100 grid points by cubic spline interpolation; arranging all grid point pressure values in spatial order to form the instantaneous pressure distribution. Combine all instantaneous distributions in time order to construct a three-dimensional matrix of pressure wave propagation (time x position x pressure value). For example, at 200 milliseconds after the valve is closed, the pressure value at a distance of 3.5 meters from the head-end can be obtained by interpolation, and the interpolation algorithm uses the standard cubic spline interpolation method.

[0110] When identifying the morphological change characteristics of the pressure wave propagation wave front, first define the wave front position: in each column of time series, find the time point at which the pressure value first drops by more than 0.5 kilopascal, and this threshold is set based on 2 times the noise level of the pressure sensor; the position corresponding to this time point is the wave front position. Scan all grid points along the pipeline axis to record the curve of the wave front position change with time. The morphological change characteristics include: the convexity of the wave front curve, the number of turning points, the curvature change rate. The specific identification method is: calculate the wave front time difference of adjacent grid points, and when the time difference change exceeds 10%, mark it as a turning point, and the 10% threshold is determined according to the waveform smoothness statistics; the convexity is determined by the angle between the connecting lines of the adjacent three points, and the angle less than 170 degrees is determined as the concave feature.

[0111] The discrete curvature algorithm is used to quantify the local bending degree of the wavefront. For each grid point on the wavefront position curve, a second-order polynomial function is fitted with the adjacent left and right two grid points (a total of five points). The curvature value calculation process is as follows: first, calculate the wavefront time difference between the center point and the second point on the left; second, calculate the wavefront time difference between the center point and the second point on the right; third, calculate the wavefront time difference between the first point on the left and the first point on the right; fourth, add the left difference value to the right difference value and subtract twice the center difference value to obtain the numerator; and fifth, divide the absolute value of the numerator by the square of the grid spacing to obtain the curvature value.

[0112] When scanning the distribution of local bending degree within the range of the pipe space, all grid points are traversed along the axial direction of the pipe. Each grid point corresponds to a curvature value, forming a curvature distribution array. The scanning process uses a sliding window mechanism: the window width is five grid points (0.5 meters), and the step size is one grid point. For each window position, the curvature value of the center point of the window is calculated and recorded. Finally, a curvature distribution graph is generated for the entire length of the pipe, with the horizontal axis representing the pipe position and the vertical axis representing the curvature value. For example, when the pipe is 10 meters long, 100 curvature data points are generated.

[0113] When marking the position points with prominent local bending degree, a dynamic threshold mechanism is set. First, calculate the average value of the curvature distribution of the entire pipe, and multiply the average value by a coefficient of 1.5 to obtain the base threshold value, which is determined based on typical leakage scenarios. At the same time, detect the gradient change of the curvature distribution, and when the gradient change rate exceeds 5% per meter, reduce the base threshold value by 20%. The marking condition is: the curvature value exceeds the dynamic threshold value and is a local maximum value (greater than the curvature values of the left and right adjacent two grid points). For example, if the average curvature is 150 milliseconds per square meter, the base threshold value is 225 milliseconds per square meter, and the curvature of a certain point is 250 milliseconds per square meter and greater than its neighbors, it is marked. The marked points record the position coordinates and curvature values.

[0114] When analyzing the spatial distribution relationship of the marked position points, cluster isolation detection is performed. All marked points are sorted by pipe axial position, and the distance between adjacent marked points is calculated. The isolation judgment condition is: when the distance between adjacent marked points is greater than 0.3 meters, it is judged to be isolated from each other, and the 0.3 meter threshold is three times the typical diameter of 0.1 meters of the pipe; when there are at least two independent isolated groups, the coordination judgment is triggered. An isolated group is defined as: the distance between points in the group is less than 0.3 meters, and the minimum distance between groups is greater than 0.3 meters. For example, if the three marked points are located at 1.2 meters, 1.5 meters, and 3.0 meters, then 1.2 meters and 1.5 meters form a group (with a distance of 0.3 meters), and 3.0 meters is an independent point, forming two isolated groups.

[0115] When generating multiple leak point collaborative back-suction judgment results, set confidence verification. When detecting two or more isolated groups, require each group to contain at least one marker point and the minimum distance between groups to be greater than 0.3 meters. The judgment result is a binary flag: output high level (5 volts) when the condition is met, otherwise output low level (0 volts). The result signal is transmitted to the joint control module through an optical coupling isolation circuit, with a response time less than 10 milliseconds. For example, when two isolated groups are detected, a 5-volt signal is output.

[0116] The implementation process includes a data verification mechanism: when the wavefront position curve is discontinuous (there is a gap of more than 5 grid points), linear interpolation is used to fill in the gap; when the curvature calculation produces a negative value, it is forced to zero; when the pipe length is less than 3 meters, the isolation distance threshold is adjusted to 10% of the pipe length. All calculations are completed within 200 milliseconds, synchronized with the valve closing action.

[0117] Dynamic threshold adjustment rule: count the number of marker points every 24 hours, and when the marker points exceed 5 in 80% of the time period, increase the coefficient from 1.5 to 1.7; when the marker points are less than 2, reduce the coefficient to 1.3. The adjustment result is written to non-volatile memory, which uses ferroelectric random access memory technology.

[0118] Wavefront detection physical principle: multiple leak points cause the pressure wave front to produce multiple depressions, the depression position corresponds to the leak point, and the depression depth is proportional to the leak rate. The maximum curvature point corresponds to the position of the wave front shape mutation, and the maximum value of the spatial isolation indicates the existence of multiple independent leak sources.

[0119] Pressure oscillation waveform acquisition synchronization mechanism: the three position pressure sensors use the same clock source, with a clock synchronization accuracy of 0.1 milliseconds, ensuring time alignment. The data buffer uses a double buffering structure to ensure continuous data acquisition without data loss.

[0120] Based on the judgment results of air back-suction risk and multiple leak point collaborative back-suction, switch the valve closing action to a two-stage mode and trigger the associated ventilation instruction, which is implemented as follows:

[0121] When receiving the air back-suction risk judgment result and the multiple leak point collaborative back-suction judgment result, obtain the two signals through a digital input interface. The air back-suction risk judgment result is represented by a 5-volt high level indicating a risk, and a 0-volt low level indicating no risk; the multiple leak point collaborative back-suction judgment result is also represented by a 5-volt high level indicating collaborative leakage, and a 0-volt low level indicating no collaborative leakage. The signals are input to the control unit through an optical isolation circuit, with a signal sampling period of 10 milliseconds. When the input signal level is stable for more than 20 milliseconds, it is determined to be a valid signal. This stable time threshold is set based on the maximum duration of electrical interference pulses.

[0122] When the air siphon risk determination result is high level, a linkage ventilation instruction is generated. The linkage ventilation instruction is a 24-volt direct-current pulse signal, the pulse width is 500 milliseconds, the pulse width is determined according to the minimum attraction time of the relay; the repetition period is 5 seconds, and the period is determined according to the response characteristics of the ventilation system. The instruction is sent to the ventilation system controller to trigger the following actions: start the explosion-proof axial flow fan, and the fan power is determined according to the pipeline volume; open the ventilation pipeline electromagnetic valve, and the valve opening degree is set to 100%; the ventilation duration is fixed to 120 seconds, which is calculated according to the time required for complete replacement of gas in the pipeline. For example, a 10 cubic meter pipeline matches a 500 watt fan, and the ventilation process maintains a 12 cubic meter per minute air volume.

[0123] When the air siphon risk high level and multiple leakage points are detected at the same time, the valve closing action is switched to a two-stage mode. The first stage control process: control the stepper motor to drive the valve rod at a speed of 30 revolutions per minute, which is 80% of the normal closing speed; close the valve from the full open position to the first opening degree position. The first opening degree position is determined by flow feedback closed loop: real-time monitoring of gas flow sensor output value, when the flow value decreases to 5% of the lower explosive limit value, stop the closing action. The lower explosive limit value is preset according to the gas composition: 5% by volume for natural gas, and 2% by volume for liquefied petroleum gas. For example, the lower explosive limit in a natural gas pipeline is 5%, and the target flow value is 0.25% (5% multiplied by 5%).

[0124] The first opening degree maintenance stage: keep the valve position unchanged, the duration is 60 seconds, which is determined according to the gas diffusion model. During this period, the gas flow value is continuously monitored, and if the flow value rises above 6% of the lower explosive limit value, the closing action is restarted until it returns to below 5% of the lower explosive limit value. The maintenance stage triggers a ventilation instruction simultaneously, and the ventilation intensity is increased to 1.5 times of the normal mode, which is realized by increasing the fan speed.

[0125] The second stage control process: after the maintenance stage ends, control the stepper motor to run in the speed reduction mode, and the speed is reduced to 10 revolutions per minute, which is one third of the speed in the first stage. The position-speed curve is planned for the closing process: the initial 10% stroke maintains 10 revolutions per minute; the middle 80% stroke is reduced to 8 revolutions per minute; and the last 10% stroke is reduced to 5 revolutions per minute. The whole closing time is controlled in the range of 15 to 25 seconds, which is adjusted according to the valve type through a preset parameter table. For example, a 20mm diameter ball valve takes 20 seconds from the first opening degree to full closure.

[0126] The implementation process sets a safety interlocking mechanism: when the flow value cannot be reduced to less than 10% of the lower explosive limit within 5 seconds during the first stage closing process, it is determined that the adjustment has failed, and the emergency closing mode is immediately switched to full-speed closing at 45 revolutions per minute; when the ventilation system failure feedback signal is valid, the execution of the two-stage mode is prohibited; when the pipeline pressure is lower than 5 kPa, the maintenance stage is automatically shortened to 30 seconds.

[0127] The two-stage mode parameter setting is based on: the first speed of 30 revolutions per minute is set at 80% of the conventional valve closing speed, which is determined through water hammer effect simulation test; the second speed of 10 revolutions per minute is one-third of the first speed, which is determined through negative pressure surge suppression test; the first opening degree maintenance time of 60 seconds is calculated based on the gas turbulent diffusion equation.

[0128] Ventilation system linkage control: the conventional ventilation instruction triggers single fan operation, and the two-stage mode triggers dual fan parallel operation. Wind volume calibration method: install a wind speed sensor at the outlet of the ventilation pipeline to adjust the fan speed in real time to stabilize the wind speed at 5 meters per second. For example, if the detected wind speed is 4.5 meters per second, increase the fan speed by 10% until it reaches 5 meters per second.

[0129] Valve position feedback calibration: closed-loop control through the voltage signal of the valve core position sensor, 0.1 volt voltage change corresponding to every 5 degree valve core angle. The control unit reads the voltage value every 10 milliseconds and converts the opening degree through linear interpolation, with an accuracy error of less than 1%. For example, the target opening degree of 30% corresponds to a voltage of 1.5 volts, and when the actual detection is 1.48 volts, continue to close 0.6 degrees.

[0130] Dynamic parameter adjustment mechanism: statistical analysis of valve closing process data every 24 hours, when the first stage closing time exceeds 8 seconds with a frequency greater than 50%, increase the first speed to 35 revolutions per minute; when the second stage induced pressure fluctuation exceeds 10 kPa with a frequency greater than 20%, reduce the second speed to 8 revolutions per minute. Adjust the parameters to write to the ferroelectric memory, with a memory erase life of 1 million times.

[0131] Explosion lower limit determination tolerance: set a buffer band of 0.1%, and the flow value within the range of 4.9% to 5.1% of the lower explosive limit is considered to meet the standard. For example, the lower explosive limit of natural gas pipeline is 5%, and the actual flow corresponds to a concentration of 4.95% to 5.05%, which triggers the opening degree maintenance.

[0132] Ventilation instruction priority setting: when there are both conventional ventilation instructions and two-stage mode ventilation instructions, the two-stage mode high-intensity ventilation is executed first. The ventilation system state feedback is returned through a 4-20 milliampere current signal, and a fault alarm is triggered when the current value is less than 6 milliampere.

[0133] The calculations involved in the embodiments are all de-dimensioned numerical calculations, and the preset parameters and threshold values in the calculations are set by a person skilled in the art according to actual conditions.

[0134] It should be noted that the application can be deployed on a device itself to realize embedded applications, or run on a PC terminal or other terminal with a user interface, thereby meeting various hardware environments and use requirements.

[0135] The above embodiments can be realized wholly or partially by software, hardware, firmware or any other combination. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server or data center to another through wireless or wired transmission. The wired transmission includes optical fiber, twisted pair, coaxial cable and the like; the wireless transmission includes infrared rays, microwaves and the like. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center and the like containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD) or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0136] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0137] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed ones can be indirect coupling or communication connection through some interfaces, devices or modules, and can be electrical, mechanical or other forms.

[0138] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0139] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0140] The functions, if realized in the form of software function modules and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0141] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0142] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A security-type intelligent diaphragm gas meter, characterized in that, include: The flow monitoring module is used to monitor the gas flow rate in real time. When the gas flow rate exceeds the preset leakage threshold, it generates a valve closing signal. The pressure gradient module is used to initiate valve closing operation in response to a valve closing signal. During the valve closing operation, it collects pressure values ​​at different locations in the pipeline and generates a pressure gradient vector based on the spatial distribution of the pressure values; it also simultaneously acquires leakage detection signals, including: The valve closing action is initiated in response to the valve closing signal; During the continuous valve closing action, pressure values ​​at different locations in the pipeline are collected by pressure sensors installed at the beginning, middle and end of the pipeline. Calculate the pressure gradient vector based on the spatial distribution of pressure values ​​at the beginning, middle, and end of the pipeline; Simultaneously acquire leak detection signals through a distributed gas sensor array; The spectral shift module is used to analyze pressure values ​​through Hilbert-Huang transform and extract the marginal spectral energy centroid shift, including: Empirical mode decomposition of the pressure value generates multiple intrinsic mode function components; Perform a Hilbert transform on each intrinsic mode function component to obtain the instantaneous frequency and instantaneous amplitude; Hilbert marginal spectrum is constructed based on instantaneous frequency and instantaneous amplitude; Calculate the energy centroid frequency of the Hilbert marginal spectrum; The energy centroid frequency is compared with the reference centroid frequency to obtain the marginal spectrum energy centroid offset. The backflow detection module is used to determine the risk of air backflow when the angle between the direction angle of the pressure gradient vector and the maximum intensity direction of the leak detection signal is less than the critical angle and the centroid offset of the marginal spectrum energy is greater than the frequency shift threshold. The leak detection module is used to calculate the local curvature distribution of the pressure wave propagation wavefront through the oscillation waveform of the pressure value. When there are at least two discrete maximum points in the local curvature distribution, it is determined that multiple leak points are co-currently drawing back. The joint control module is used to switch the valve closing action to a two-stage mode and trigger a joint ventilation command based on the judgment results of the risk of backflow of air and the coordinated backflow of multiple leak points.

2. The security-type intelligent diaphragm gas meter according to claim 1, characterized in that, Real-time monitoring of gas flow rate; when the gas flow rate exceeds a preset leakage threshold, a valve shut-off signal is generated, including: The gas flow rate is obtained in real time through a diaphragm metering device; Compare the gas flow rate value with the preset leakage threshold; When the gas flow rate exceeds the preset leakage threshold, a valve shut-off signal is generated.

3. A security-type intelligent diaphragm gas meter according to claim 2, characterized in that, in, If the gas flow rate exceeds the preset leakage threshold for a preset duration, an operation to generate a valve closing signal will be executed.

4. A security-type intelligent diaphragm gas meter according to claim 1, characterized in that, Perform a Hilbert transform on each intrinsic mode function component to obtain the instantaneous frequency and instantaneous amplitude, including: Construct analytical signals from the intrinsic mode function components; Extract the phase angle function of the analytical signal, calculate the first derivative of the phase angle function with respect to time, and divide the value of the first derivative of the phase angle function with respect to time by twice pi to obtain the instantaneous frequency; Calculate the sum of squares of the real and imaginary parts of the analytic signal, and perform an arithmetic square root operation on the sum of squares to obtain the instantaneous amplitude.

5. A security-type intelligent diaphragm gas meter according to claim 1, characterized in that, When the angle between the direction angle of the pressure gradient vector and the maximum intensity orientation of the leak detection signal is less than the critical angle and the centroid shift of the marginal spectral energy is greater than the frequency shift threshold, an air backflow risk is determined, including: Calculate the direction angle of the pressure gradient vector; Determine the location of the maximum intensity of the leak detection signal; Calculate the angle between the direction angle and the azimuth of maximum intensity; Obtain the centroid offset of the marginal spectral energy; Compare the included angle with the critical angle, and compare the marginal spectral energy centroid shift with the frequency shift threshold; When the included angle is less than the critical angle and the centroid shift of the marginal spectral energy is greater than the frequency shift threshold, an air backflow risk assessment result is generated.

6. A security-type intelligent diaphragm gas meter according to claim 1, characterized in that, The local curvature distribution of the pressure wave propagation front is calculated by analyzing the oscillating waveform of the pressure value. When there are at least two discrete maxima in the local curvature distribution, it is determined that multiple leakage points are co-currently drawing back, including: Obtain the oscillation waveform of pressure values ​​at different locations in the pipeline; Reconstructing the propagation process of pressure waves in the pipeline based on oscillating waveforms; Identify the morphological changes of the propagation wavefront of pressure waves; Quantify the degree of local bending of the wavefront based on morphological change characteristics; Scan the distribution of localized curvature within the spatial range of the pipeline; Mark the locations where the degree of local curvature is prominent; Analyze the spatial distribution relationship of the marked location points; When there are at least two mutually isolated marker locations, multiple leak point collaborative backflow determination results are generated.

7. A security-type intelligent diaphragm gas meter according to claim 1, characterized in that, Based on the assessment of the risk of backdraft and the coordinated backdraft from multiple leak points, the valve closing action is switched to a two-stage mode and a linked ventilation command is triggered, including: Receive the results of the air backflow risk assessment and the results of the coordinated backflow assessment from multiple leak points; When there is a risk of backdraft, a coordinated ventilation command is generated to dilute the residual gas in the duct. When there is a risk of backflow of air and backflow from multiple leak points, the valve closing action is switched to a two-stage mode.

8. A security-type intelligent diaphragm gas meter according to claim 7, characterized in that, The two-stage mode is as follows: in the first stage, the valve is closed to the first opening degree at the first speed, and the first opening degree maintains the gas flow rate value below the lower explosive limit of the mixed gas; in the second stage, the valve is closed to the fully closed state at the second speed, which is lower than the first speed.

Images