A gas pressure stabilizing and odorant trace concentration monitoring system for gas pipeline

Abstract

This invention relates to the field of pipeline monitoring technology, specifically a gas pressure stability and odorant trace concentration monitoring system for gas pipelines. The system includes a pressure decomposition module, an instruction generation module, a coarse concentration measurement module, a fine concentration measurement module, a signal early warning module, and a loss estimation module. It performs empirical mode decomposition on the raw pressure fluctuation time-series data, using the instantaneous amplitude of the high-frequency intrinsic mode component as the gas pressure disturbance characteristic value. It then queries a disturbance-compensation mapping table to obtain the feedforward compensation coefficient and generates a gas pressure compensation instruction. The system uses stable airflow to identify characteristic peaks in the raw spectral response signal to obtain the coarse odorant concentration. Next, it obtains a baseline drift function through polynomial fitting, corrects it point-by-point to obtain the trace concentration monitoring value, compares the monitoring value with a safety threshold to output an anomaly warning, and stores it in association with a timestamp to form a concentration curve for predicting the odorant loss rate. This invention can improve the efficiency of odorant trace concentration monitoring.

Description

Technical Field

[0001] This invention relates to the field of pipeline monitoring technology, and in particular to a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines. Background Technology

[0002] In the field of gas pipeline operation monitoring, existing technologies for processing gas pressure fluctuations in gas pipeline networks mostly adopt traditional filtering or fixed compensation methods. These methods cannot accurately perform modal decomposition and feature extraction on the time series data of pressure fluctuations, making it difficult to capture high-frequency gas pressure disturbance characteristics within the pipeline network. Consequently, the generation of gas pressure compensation commands lacks accurate disturbance data support, resulting in delayed compensation operations and poor adaptability. As a result, the gas pressure in the gas pipeline network is always difficult to maintain a stable state, introducing significant airflow interference errors for subsequent monitoring of odorant concentration.

[0003] Existing odorant concentration monitoring technologies mostly perform only coarse analysis, failing to accurately fit and subtract baseline drift issues in spectral response signals. They also ignore the influence of environmental factors such as temperature and pressure on concentration detection, making it impossible to accurately monitor trace odorant concentrations. Furthermore, the monitoring data lacks time-series correlation analysis, making it difficult to effectively fit concentration change curves and predict odorant depletion rates. This results in a lack of timely and accurate early warnings for concentration anomalies and a failure to provide scientific rate references for odorant replenishment in pipelines. Overall, the level of intelligence and accuracy in monitoring and maintenance is low. Therefore, improving the efficiency of odorant concentration monitoring has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines, characterized in that the system includes a gas pressure decomposition module, an instruction generation module, a coarse concentration measurement module, a fine concentration measurement module, a signal early warning module, and a loss estimation module, wherein:

[0005] The pressure decomposition module is used to perform empirical mode decomposition on the original pressure fluctuation time series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and use the instantaneous amplitude of the high-frequency intrinsic mode components as the pressure disturbance characteristic value of the gas pipeline network.

[0006] The instruction generation module is used to query a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value, obtain the feedforward compensation coefficient of the gas pipeline network, and encode the feedforward compensation coefficient into the gas pressure compensation instruction of the gas pipeline network.

[0007] The concentration coarse measurement module is used to receive the stable airflow generated after the gas pressure compensation command is executed, and based on the stable airflow, to identify the characteristic peaks of the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network.

[0008] The concentration precision measurement module is used to perform polynomial fitting on the original spectral response signal to obtain the baseline drift function of the gas pipeline network, and to subtract the coarse concentration value of the odorant point by point according to the baseline drift function to obtain the trace concentration monitoring value of the odorant in the gas pipeline network.

[0009] The signal warning module is used to compare and analyze the monitoring value of the trace concentration of the odorant with a preset concentration safety threshold range to obtain an abnormal warning signal of the gas pipeline network.

[0010] The loss estimation module is used to associate and store the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, so as to predict the odorant loss rate of the gas pipeline network.

[0011] In a preferred embodiment, when the pressure decomposition module performs empirical mode decomposition on the raw pressure fluctuation time-series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and uses the instantaneous amplitude of the high-frequency intrinsic mode components as the pressure disturbance characteristic value of the gas pipeline network, it is specifically used for:

[0012] The raw pressure fluctuation time series data of the node under test in the gas pipeline network is obtained within a continuous sampling period, and the raw pressure fluctuation time series data is filtered by a sliding window to obtain the smooth pressure time series data of the gas pipeline network.

[0013] Local extremum points are detected in the smoothed pressure time series data to obtain the local maxima and local minima of the smoothed pressure time series data.

[0014] By fitting the local maxima and local minima, the upper and lower envelopes of the gas pipeline network are obtained.

[0015] The upper and lower envelopes are averaged to obtain the local mean curves of the gas pipeline network, and the local mean curves are removed to obtain the candidate intrinsic mode components of the gas pipeline network.

[0016] The candidate intrinsic mode components are screened and a termination condition is determined to obtain the high-frequency intrinsic mode components of the gas pipeline network.

[0017] The high-frequency intrinsic mode components are subjected to Hilbert transform to obtain the instantaneous frequency of the high-frequency intrinsic mode components, and the amplitude corresponding to the instantaneous frequency is used as the gas pressure disturbance characteristic value of the gas pipeline network.

[0018] In a preferred embodiment, when the instruction generation module executes the query of a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value to obtain the feedforward compensation coefficient of the gas pipeline network, and encodes the feedforward compensation coefficient into a gas pressure compensation instruction for the gas pipeline network, it is specifically used for:

[0019] Membership analysis is performed on the pressure disturbance characteristic values ​​to determine the target disturbance level range to which the pressure disturbance characteristic values ​​belong;

[0020] Based on the target disturbance level range, the corresponding initial value of the feedforward compensation coefficient is extracted from the preset disturbance-compensation mapping table and used as the compensation coefficient to be corrected for the gas pipeline network.

[0021] The difference between the pressure disturbance characteristic value and the interval endpoint value of the target disturbance level interval is compared to obtain the correction weight of the compensation coefficient to be corrected.

[0022] Based on the correction weight, the compensation coefficient to be corrected is linearly adjusted to obtain the feedforward compensation coefficient of the air pressure disturbance characteristic value.

[0023] The feedforward compensation coefficient is filled into the data field of the instruction frame in the gas pipeline network to obtain the gas pressure compensation instruction of the gas pipeline network.

[0024] In a preferred embodiment, when the concentration coarse measurement module executes the stable airflow generated after receiving the pressure compensation command, and based on the stable airflow, performs characteristic peak identification on the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network, it is specifically used for:

[0025] By applying the pressure compensation command and monitoring the gas flow data of the gas pipeline network in real time, a stable gas flow of the gas pipeline network is obtained.

[0026] The original spectral response signal of the gas pipeline network is parsed to extract a continuous spectral data frame sequence corresponding to the period of stable airflow generation.

[0027] Noise estimation is performed on the continuous spectral data frame sequence to obtain the local baseline noise level of the original spectral response signal, and the peak detection threshold of the original spectral response signal is set according to the local baseline noise level.

[0028] Based on the peak detection threshold, the candidate characteristic peaks of the original spectral response signal are obtained by traversing the continuous spectral data frame sequence.

[0029] The peak wavelength position and peak height of the candidate characteristic peaks are extracted, and combined with the preset standard odorant characteristic spectral library, the effective characteristic peaks of the original spectral response signal are obtained;

[0030] The effective characteristic peaks are quantitatively inverted to obtain the rough concentration value of the odorant in the gas pipeline network.

[0031] In a preferred embodiment, when the concentration precision measurement module performs polynomial fitting on the original spectral response signal of the gas pipeline network to obtain the baseline drift function of the gas pipeline network, and subtracts the coarse concentration value of the odorant point by point according to the baseline drift function to obtain the trace concentration monitoring value of the odorant in the gas pipeline network, it is specifically used for:

[0032] The discrete data points in the original spectral response signal are used as the baseline fitting sample points of the gas pipeline network.

[0033] Set the order of the polynomial fitting, use wavenumber as the independent variable and absorbance as the dependent variable, and perform least squares fitting on the baseline fitting sample points to obtain the polynomial coefficient set.

[0034] Based on the set of polynomial coefficients, a baseline drift function for the gas pipeline network is constructed. The baseline drift function takes wavenumber as input and outputs the corresponding baseline absorbance value.

[0035] Obtain the wavenumber sequence corresponding to the crude concentration value of the odorant, and substitute the wavenumber sequence into the baseline drift function to obtain the baseline absorbance shift corresponding to the wavenumber sequence.

[0036] According to the preset conversion rules, the baseline absorbance offset is converted into a baseline concentration offset;

[0037] The difference between the coarse concentration value of the odorant and the baseline concentration offset is compared to obtain the trace concentration monitoring value of the odorant in the gas pipeline network.

[0038] In a preferred embodiment, when the concentration precision measurement module performs the conversion of the baseline absorbance offset into a baseline concentration offset according to a preset conversion rule, it is specifically used for:

[0039] The current ambient temperature, current ambient pressure, reference temperature, reference pressure, temperature influence coefficient, pressure influence coefficient, and instrument correction factor of the gas pipeline network are obtained.

[0040] The concentration absorption peak of the odorant is obtained by performing deconvolution fitting on the crude concentration value of the odorant.

[0041] Peak shape fitting is performed on the concentration absorption peak to obtain the peak center wavenumber and peak position weighting coefficient of the concentration absorption peak;

[0042] The baseline concentration offset of the odorant's rough concentration value is calculated according to the preset conversion rules.

[0043] In a preferred embodiment, the formula for calculating the baseline concentration offset is:

[0044] ;

[0045] in, This indicates the baseline concentration offset. This represents the instrument's correction factor. This represents the baseline absorbance offset. Indicates the first The aforementioned peak position weighting coefficients. Indicates the first The peak center wavenumbers mentioned above This indicates the current wavenumber position of the wavenumber sequence. This represents the preset peak shape attenuation coefficient. This represents a preset background constant. This represents the temperature influence coefficient. This indicates the current ambient temperature value. This indicates the reference temperature. This represents the pressure influence coefficient. This indicates the current environmental pressure. This indicates the reference pressure. Represents the natural constant.

[0046] In a preferred embodiment, when the signal warning module performs a comparison and analysis of the odorant trace concentration monitoring value with a preset concentration safety threshold range to obtain an abnormal warning signal for the gas pipeline network, it is specifically used for:

[0047] Obtain historical monitoring values ​​of the trace concentration of the odorant, and construct a sliding window sequence for monitoring the concentration of the trace concentration of the odorant;

[0048] The concentration monitoring sliding window sequence is processed by moving average to obtain the window smoothed concentration value of the odorant trace concentration monitoring value;

[0049] The window smoothed concentration value is compared with a preset concentration safety threshold range. When the window smoothed concentration value is lower than the lower limit of the concentration safety threshold range, a low concentration anomaly marker is triggered, and the deviation between the window smoothed concentration value and the lower limit value is recorded.

[0050] When the window-smoothed concentration value is higher than the upper limit of the concentration safety threshold range, a high concentration anomaly marker is triggered, and the deviation between the window-smoothed concentration value and the upper limit value is recorded.

[0051] An abnormality warning signal for the gas pipeline network is generated based on the abnormality marker and the deviation amplitude.

[0052] In a preferred embodiment, when the loss estimation module performs the process of associating and storing the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for:

[0053] Add a timestamp of the acquisition time to the odorant trace concentration monitoring value to obtain a time-stamped concentration data tuple of the odorant trace concentration monitoring value;

[0054] According to the time sequence, the time-stamped concentration data tuples are inserted into the circular storage buffer of the gas pipeline network to obtain the concentration time sequence data chain of the odorant trace concentration monitoring value;

[0055] The concentration time-series data chain is periodically scanned to obtain the current periodic concentration subsequence of the odorant trace concentration monitoring value;

[0056] Outlier removal is performed on the current periodic concentration subsequence to obtain a clean periodic concentration subsequence of the odorant trace concentration monitoring values;

[0057] By performing line fitting on the discrete data points in the clean periodic concentration subsequence, periodic concentration change segments of the odorant trace concentration monitoring values ​​are obtained.

[0058] By splicing the periodic concentration change segments together in chronological order, the odorant concentration change curve of the gas pipeline network is obtained, in order to predict the odorant consumption rate of the gas pipeline network.

[0059] In a preferred embodiment, when the loss estimation module performs the step of splicing the periodic concentration change segments in chronological order to obtain the odorant concentration change curve of the gas pipeline network in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for:

[0060] Identify the monotonically decreasing segment in the concentration change curve of the odorant, and extract the starting point concentration value, starting point time, ending point concentration value, and ending point time of the monotonically decreasing segment;

[0061] The concentration difference between the starting point concentration value and the ending point concentration value is calculated, as well as the time interval between the starting point time and the ending point time.

[0062] The instantaneous loss rate of the monotonically decreasing segment is determined based on the concentration difference and the time interval.

[0063] The instantaneous loss rate with the largest value is taken as the peak loss rate, and the instantaneous loss rate with the smallest value is taken as the trough loss rate.

[0064] The average of the peak loss rate, the trough loss rate, and the instantaneous loss rate is used as the predicted odorant loss rate of the gas pipeline network.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. This invention, through a multi-module collaborative intelligent technical architecture, achieves a precise upgrade in the monitoring of gas pipeline pressure stability and odorant concentration. In the pressure control stage, it relies on empirical mode decomposition to extract pressure disturbance feature values ​​and combines them with a preset mapping table to generate precise feedforward compensation commands. This enables rapid response to pressure fluctuations in the gas pipeline and targeted compensation, effectively improving the timeliness and accuracy of gas pipeline pressure stability control and laying a stable airflow foundation for odorant concentration monitoring. Regarding odorant concentration monitoring, a dual-layer detection method combining coarse and fine measurements is adopted. First, coarse concentration measurement is completed through feature peak identification. Then, a baseline drift function is constructed through polynomial fitting, and the coarse measurement value is subtracted point by point. Simultaneously, environmental parameters are combined to correct the concentration offset, significantly improving the accuracy of odorant trace concentration monitoring and achieving refined perception of odorant concentration in the gas pipeline network.

[0067] 2. This invention, by constructing a concentration monitoring sliding window sequence and performing moving average processing, can accurately identify abnormal fluctuations in odorant concentration and generate corresponding early warning signals, improving the sensitivity and reliability of abnormal concentration early warning in gas pipeline networks and facilitating timely handling of potential risks by staff. The loss estimation module constructs a concentration change curve by correlating concentration monitoring values ​​with timestamps, accurately extracts characteristic segments of the curve, and calculates the loss rate, effectively improving the scientific rigor and accuracy of odorant loss rate prediction. This provides data support for the replenishment and management of odorants in gas pipeline networks. Furthermore, each module can be implemented independently and supports flexible calling and expansion, improving the overall adaptability and operational efficiency of the system and ensuring the stable and safe operation of gas pipeline networks. Attached Figure Description

[0068] Figure 1 This invention provides a system architecture diagram of a gas pressure stabilization and odorant trace concentration monitoring system for a gas pipeline, according to an embodiment of the present invention.

[0069] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0072] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0073] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0074] In practice, the server-side equipment deployed in a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines may consist of one or more devices. This system can be implemented as a business instance, a virtual machine, or hardware equipment. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node, providing gas pressure stabilization and odorant trace concentration monitoring services to various user terminals. Alternatively, it can also be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage each user terminal. Alternatively, the gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines can also be implemented as a server consisting of numerous identical or different types of hardware devices, with one or more hardware devices set up to provide each user terminal with a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines.

[0075] In terms of implementation, a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines and its user terminal are mutually compatible. Specifically, if the gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines is implemented as an application installed on a cloud service platform, then the user terminal acts as a client establishing a communication connection with that application; or if the system is implemented as a website, then the user terminal acts as a webpage; or if the system is implemented as a cloud service platform, then the user terminal acts as a mini-program within an instant messaging application.

[0076] like Figure 1 The diagram shown is a system architecture diagram of a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines provided by an embodiment of the present invention.

[0077] The gas pressure stabilization and odorant trace concentration monitoring system 100 for gas pipelines described in this invention can be installed on a cloud server. In terms of implementation, it can be used as one or more service devices, or as an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, the gas pressure stabilization and odorant trace concentration monitoring system 100 for gas pipelines may include a gas pressure decomposition module 101, an instruction generation module 102, a coarse concentration measurement module 103, a fine concentration measurement module 104, a signal early warning module 105, and a loss estimation module 106. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0078] In this embodiment of the invention, in a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines, each of the above-mentioned modules can be implemented independently and can be invoked by other modules. Invocation here can be understood as one module connecting to multiple modules of another type and providing corresponding services to the connected modules. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines provided by this embodiment of the invention allows for adjustment of the applicable scope of the system architecture without modifying the program code, through adding modules and directly invoking them. This enables cluster-based horizontal expansion, achieving the goal of quickly and flexibly expanding the gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines. In practical applications, the above modules can be set in the same device or different devices, or they can be set in a virtual device, such as a service instance in a cloud server.

[0079] The following describes, with reference to specific embodiments, each component and its specific workflow of a gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines:

[0080] The pressure decomposition module 101 is used to perform empirical mode decomposition on the original pressure fluctuation time series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and use the instantaneous amplitude of the high-frequency intrinsic mode components as the pressure disturbance characteristic value of the gas pipeline network.

[0081] In this embodiment of the invention, when the gas pressure decomposition module performs empirical mode decomposition on the raw pressure fluctuation time-series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and uses the instantaneous amplitude of the high-frequency intrinsic mode components as the gas pressure disturbance characteristic value of the gas pipeline network, it is specifically used for:

[0082] The raw pressure fluctuation time series data of the node under test in the gas pipeline network is obtained within a continuous sampling period, and the raw pressure fluctuation time series data is filtered by a sliding window to obtain the smooth pressure time series data of the gas pipeline network.

[0083] Local extremum points are detected in the smoothed pressure time series data to obtain the local maxima and local minima of the smoothed pressure time series data.

[0084] By fitting the local maxima and local minima, the upper and lower envelopes of the gas pipeline network are obtained.

[0085] The upper and lower envelopes are averaged to obtain the local mean curves of the gas pipeline network, and the local mean curves are removed to obtain the candidate intrinsic mode components of the gas pipeline network.

[0086] The candidate intrinsic mode components are screened and a termination condition is determined to obtain the high-frequency intrinsic mode components of the gas pipeline network.

[0087] The high-frequency intrinsic mode components are subjected to Hilbert transform to obtain the instantaneous frequency of the high-frequency intrinsic mode components, and the amplitude corresponding to the instantaneous frequency is used as the gas pressure disturbance characteristic value of the gas pipeline network.

[0088] Continuous sampling of pressure data is performed on preset test nodes in the gas pipeline network. The original pressure fluctuation time series data is formed by organizing the sampling time sequence. A fixed-length time window is selected and slides point by point on the original pressure fluctuation time series data. The average pressure data in each window is calculated and the calculated average value is used to replace the original pressure data in the window. This completes the sliding window filtering process of the original pressure fluctuation time series data, and finally the smooth pressure time series data of the gas pipeline network is obtained.

[0089] Iterate through all data points in the smoothed pressure time series data, and compare the pressure value of each data point with the pressure values ​​of its adjacent data points one by one. If the pressure value of the data point is greater than the pressure values ​​of the adjacent data points, it is determined to be a local maximum point. If the pressure value of the data point is less than the pressure values ​​of the adjacent data points, it is determined to be a local minimum point. This completes the detection of local extreme points in the smoothed pressure time series data, and obtains the corresponding local maximum and local minimum points.

[0090] All detected local maxima points are arranged in chronological order. Interpolation is used to complete the data of the arranged local maxima points and connect them into a continuous curve to form the upper envelope of the gas pipeline network. In the same way, all local minima points are arranged and interpolated to complete the data and connect them into a continuous curve to form the lower envelope of the gas pipeline network.

[0091] The pressure values ​​corresponding to the upper and lower envelopes at the same time node are summed, and the sum is divided by two to complete the mean value processing of the two envelopes, forming a local mean curve of the gas pipeline network with the same time dimension as the envelopes. The pressure values ​​at the same time node of the smoothed pressure time series data are subtracted from the values ​​corresponding to the local mean curve, and the resulting difference sequence is used as the candidate intrinsic mode components of the gas pipeline network.

[0092] A fixed screening termination criterion is set, and candidate intrinsic mode components are substituted into the criterion for compliance verification. If the candidate intrinsic mode component meets the criterion, it is directly identified as the high-frequency intrinsic mode component of the gas pipeline network. If it does not meet the criterion, the envelope fitting, meanization and difference calculation operations are repeated for the candidate intrinsic mode component until the processed component meets the screening termination criterion, and finally the high-frequency intrinsic mode component of the gas pipeline network is obtained.

[0093] Hilbert transform is performed on the determined high-frequency intrinsic mode components to convert the time-domain signal of the high-frequency intrinsic mode components into a frequency-domain signal. The instantaneous frequency corresponding to each time node is extracted from the converted frequency-domain signal, and then the amplitude value corresponding to each instantaneous frequency in the frequency-domain signal is extracted. This amplitude value is directly used as the gas pressure disturbance characteristic value of the gas pipeline network.

[0094] The beneficial effects of this implementation process are that it effectively removes noise from the original pressure fluctuation time series data through the specific operation of sliding window filtering, ensuring the accuracy of subsequent pressure fluctuation characteristic analysis from the data source. Local extreme point detection is achieved through point-by-point comparison, accurately capturing the extreme fluctuation characteristics of smoothed pressure time series data. Combined with interpolation, the envelope is fitted and connected, allowing the upper and lower envelopes to completely match the fluctuation trend of the smoothed pressure time series data. The local mean curve obtained after meanization accurately reflects the baseline trend of pressure fluctuations. The candidate intrinsic mode components obtained after removing this curve can truly reflect the essential characteristics of pressure fluctuations in the gas pipeline network. Through screening... The repeated verification of termination conditions ensures that the extracted high-frequency intrinsic mode components fully meet the requirements of gas pressure disturbance feature analysis. Then, the precise conversion from the time domain to the frequency domain is achieved through Hilbert transform, which can accurately extract the amplitude corresponding to the instantaneous frequency from the high-frequency intrinsic mode components as the gas pressure disturbance feature value. Every step of the entire process revolves around the precise extraction of gas pressure disturbance features, forming a standardized and refined processing flow. This provides solid and effective data support for the subsequent command generation module to obtain accurate gas pressure disturbance data and generate appropriate gas pressure compensation commands, making the feature identification of gas pressure disturbances in gas pipeline networks more targeted, accurate, and professional.

[0095] The instruction generation module 102 is used to query a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value, obtain the feedforward compensation coefficient of the gas pipeline network, and encode the feedforward compensation coefficient into the gas pressure compensation instruction of the gas pipeline network.

[0096] In this embodiment of the invention, when the instruction generation module executes the query of a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value to obtain the feedforward compensation coefficient of the gas pipeline network, and encodes the feedforward compensation coefficient into a gas pressure compensation instruction for the gas pipeline network, it is specifically used for:

[0097] Membership analysis is performed on the pressure disturbance characteristic values ​​to determine the target disturbance level range to which the pressure disturbance characteristic values ​​belong;

[0098] Based on the target disturbance level range, the corresponding initial value of the feedforward compensation coefficient is extracted from the preset disturbance-compensation mapping table and used as the compensation coefficient to be corrected for the gas pipeline network.

[0099] The difference between the pressure disturbance characteristic value and the interval endpoint value of the target disturbance level interval is compared to obtain the correction weight of the compensation coefficient to be corrected.

[0100] Based on the correction weight, the compensation coefficient to be corrected is linearly adjusted to obtain the feedforward compensation coefficient of the air pressure disturbance characteristic value.

[0101] The feedforward compensation coefficient is filled into the data field of the instruction frame in the gas pipeline network to obtain the gas pressure compensation instruction of the gas pipeline network.

[0102] Membership analysis is performed on the characteristic value of the pressure disturbance. Based on the characteristic value range of each preset disturbance level interval, the characteristic value of the pressure disturbance is matched with each interval range one by one to identify the disturbance level interval that the characteristic value of the pressure disturbance completely corresponds to, and the interval is determined as the target disturbance level interval.

[0103] Based on the determined target disturbance level range, the preset disturbance-compensation mapping table is retrieved, and the initial value of the feedforward compensation coefficient that uniquely corresponds to the target disturbance level range is searched in the mapping table. This initial value is then directly used as the compensation coefficient to be corrected for the gas pipeline network.

[0104] Obtain the upper and lower limit endpoints of the target disturbance level range, calculate the difference between the pressure disturbance characteristic value and the upper and lower limit endpoints of the range, and calculate the correction weight used to adjust the compensation coefficient to be corrected based on the numerical ratio of the two differences.

[0105] The compensation coefficient to be corrected is linearly operated on with the calculated correction weight. This operation is used to precisely adjust the value of the compensation coefficient to be corrected. The adjusted value is the feedforward compensation coefficient that adapts to the characteristic value of the air pressure disturbance.

[0106] The pre-set command frame structure of the gas pipeline network is retrieved, and the determined feedforward compensation coefficient is completely filled into the specified position of the data field of the command frame. After the data filling of the command frame is completed, a gas pressure compensation command for the gas pipeline network that can be directly issued and executed is formed.

[0107] The beneficial effects of this implementation process are that it accurately locates the target disturbance level range of gas pressure disturbance characteristic values ​​through membership analysis, providing a precise range basis for the selection of compensation coefficients. The initial values ​​extracted from the disturbance-compensation mapping table ensure the basic adaptability of the compensation coefficients. The correction weight is calculated by combining the difference between the characteristic value and the interval endpoint value, giving the adjustment of the compensation coefficients precise data support. The fine optimization of the compensation coefficients to be corrected is achieved through linear adjustment, so that the final feedforward compensation coefficients are highly matched with the actual gas pressure disturbance characteristics. Then, the compensation coefficients are standardized and filled into the data field of the instruction frame to form a gas pressure compensation instruction, ensuring the standardization and executability of the instruction. The entire process is progressive, realizing the accurate acquisition of compensation coefficients and the standardized generation of instructions, so that the gas pressure compensation instruction can accurately match the actual gas pressure disturbance situation of the gas pipeline network, providing a precise execution basis for subsequent gas pressure stabilization regulation.

[0108] The concentration coarse measurement module 103 is used to receive the stable airflow generated after the gas pressure compensation command is executed, and based on the stable airflow, to identify the characteristic peaks of the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network.

[0109] In this embodiment of the invention, when the concentration coarse measurement module executes the stable airflow generated after receiving the gas pressure compensation command, and based on the stable airflow, performs characteristic peak identification on the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network, it is specifically used for:

[0110] By applying the pressure compensation command and monitoring the gas flow data of the gas pipeline network in real time, a stable gas flow of the gas pipeline network is obtained.

[0111] The original spectral response signal of the gas pipeline network is parsed to extract a continuous spectral data frame sequence corresponding to the period of stable airflow generation.

[0112] Noise estimation is performed on the continuous spectral data frame sequence to obtain the local baseline noise level of the original spectral response signal, and the peak detection threshold of the original spectral response signal is set according to the local baseline noise level.

[0113] Based on the peak detection threshold, the candidate characteristic peaks of the original spectral response signal are obtained by traversing the continuous spectral data frame sequence.

[0114] The peak wavelength position and peak height of the candidate characteristic peaks are extracted, and combined with the preset standard odorant characteristic spectral library, the effective characteristic peaks of the original spectral response signal are obtained;

[0115] The effective characteristic peaks are quantitatively inverted to obtain the rough concentration value of the odorant in the gas pipeline network.

[0116] The gas pressure compensation command is sent to the gas pressure regulation execution unit of the gas pipeline network. The execution unit adjusts the gas pressure of the pipeline network according to the command parameters. At the same time, the gas flow monitoring sensor continuously collects airflow data such as airflow velocity and airflow stability in the gas pipeline network until the monitored airflow data reaches the preset stability judgment standard. At this time, the airflow formed in the pipeline network is the stable airflow of the gas pipeline network.

[0117] The raw spectral response signals collected from the gas pipeline network are analyzed in a structured manner, and independent spectral data frames are split according to the timestamp of signal acquisition. At the same time, the start and end timestamps of stable airflow are retrieved. All data frames within the time range are selected from the split spectral data frames and arranged in chronological order to form a continuous spectral data frame sequence of the gas pipeline network.

[0118] Frame-by-frame noise statistics are performed on the continuous spectral data frame sequence. The signal fluctuation amplitude and fluctuation frequency in the region without characteristic peaks in each frame of spectral data are calculated. The local baseline noise level of the original spectral response signal is obtained by combining the statistical results of multiple frames. Based on the value of the local baseline noise level, a fixed value higher than the value is set as the peak detection threshold of the original spectral response signal.

[0119] Using the set peak detection threshold as the criterion, each spectral data point in the continuous spectral data frame sequence is traversed, and the signal value of the data point is compared with the peak detection threshold. All spectral data points with signal values ​​higher than the peak detection threshold are selected, and these data points are integrated according to the wavelength dimension to form a continuous peak structure. This peak structure is the candidate characteristic peak of the original spectral response signal.

[0120] Feature parameters are extracted from all candidate feature peaks obtained through screening, and the peak wavelength position and peak height value corresponding to each candidate feature peak are accurately obtained. At the same time, the preset standard odorant feature spectrum library is retrieved, and the peak wavelength position and peak height value of the candidate feature peaks are accurately matched with the feature peak parameters of the standard odorant in the spectrum library. Candidate feature peaks that match the standard parameters are screened out, and these feature peaks are the effective feature peaks of the original spectral response signal.

[0121] Based on the fundamental principle of quantitative spectral analysis, the peak height of the effective characteristic peak is substituted into the correlation analysis system between the odorant concentration and the peak height of the characteristic peak. By performing concentration conversion and numerical calibration on the peak height of the effective characteristic peak, the quantitative inversion of the effective characteristic peak is completed, and finally the rough concentration value of the odorant in the gas pipeline network is obtained.

[0122] The beneficial effects are as follows: This implementation process first achieves accurate acquisition of stable airflow by executing air pressure compensation commands and monitoring airflow in real time, eliminating the interference of air pressure fluctuations on spectral detection from the source and ensuring the basic accuracy of subsequent concentration detection. By parsing data frames and extracting time periods to accurately match the spectral data corresponding to stable airflow, concentration detection is carried out only for signals in the effective detection period, improving the targeting of data processing. Combining noise estimation to set peak detection thresholds can effectively eliminate noise interference and accurately screen candidate characteristic peaks. By matching parameters with the standard odorant characteristic spectral library, the effective characteristic peaks are accurately identified, avoiding the influence of invalid peak shapes on concentration detection. Finally, quantitative inversion completes the conversion of effective characteristic peaks to coarse concentration values, forming a standardized coarse odorant detection process. This provides accurate and effective basic concentration data for subsequent fine detection, and improves the overall accuracy and effectiveness of odorant concentration detection.

[0123] The concentration precision measurement module 104 is used to perform polynomial fitting on the original spectral response signal to obtain the baseline drift function of the gas pipeline network, and to subtract the coarse concentration value of the odorant point by point according to the baseline drift function to obtain the trace concentration monitoring value of the odorant in the gas pipeline network.

[0124] In this embodiment of the invention, when the concentration precision measurement module performs polynomial fitting on the original spectral response signal of the gas pipeline network to obtain the baseline drift function of the gas pipeline network, and subtracts the coarse concentration value of the odorant point by point according to the baseline drift function to obtain the trace concentration monitoring value of the odorant in the gas pipeline network, it is specifically used for:

[0125] The discrete data points in the original spectral response signal are used as the baseline fitting sample points of the gas pipeline network.

[0126] Set the order of the polynomial fitting, use wavenumber as the independent variable and absorbance as the dependent variable, and perform least squares fitting on the baseline fitting sample points to obtain the polynomial coefficient set.

[0127] Based on the set of polynomial coefficients, a baseline drift function for the gas pipeline network is constructed. The baseline drift function takes wavenumber as input and outputs the corresponding baseline absorbance value.

[0128] Obtain the wavenumber sequence corresponding to the crude concentration value of the odorant, and substitute the wavenumber sequence into the baseline drift function to obtain the baseline absorbance shift corresponding to the wavenumber sequence.

[0129] According to the preset conversion rules, the baseline absorbance offset is converted into a baseline concentration offset;

[0130] The difference between the coarse concentration value of the odorant and the baseline concentration offset is compared to obtain the trace concentration monitoring value of the odorant in the gas pipeline network.

[0131] When the concentration precision measurement module performs the conversion of the baseline absorbance offset into a baseline concentration offset according to a preset conversion rule, it is specifically used for:

[0132] The current ambient temperature, current ambient pressure, reference temperature, reference pressure, temperature influence coefficient, pressure influence coefficient, and instrument correction factor of the gas pipeline network are obtained.

[0133] The concentration absorption peak of the odorant is obtained by performing deconvolution fitting on the crude concentration value of the odorant.

[0134] Peak shape fitting is performed on the concentration absorption peak to obtain the peak center wavenumber and peak position weighting coefficient of the concentration absorption peak;

[0135] The baseline concentration offset of the odorant's rough concentration value is calculated according to the preset conversion rules.

[0136] The formula for calculating the baseline concentration offset is:

[0137] ;

[0138] in, This indicates the baseline concentration offset. This represents the instrument's correction factor. This represents the baseline absorbance offset. Indicates the first The aforementioned peak position weighting coefficients. Indicates the first The peak center wavenumbers mentioned above This indicates the current wavenumber position of the wavenumber sequence. This represents the preset peak shape attenuation coefficient. This represents a preset background constant. This represents the temperature influence coefficient. This indicates the current ambient temperature value. This indicates the reference temperature. This represents the pressure influence coefficient. This indicates the current environmental pressure. This indicates the reference pressure. Represents the natural constant.

[0139] All discrete data points in the original spectral response signal of the gas pipeline network are extracted and used as baseline fitting sample points for baseline fitting of the gas pipeline network, providing a complete sample data foundation for the subsequent construction of the baseline drift function.

[0140] A fixed order for polynomial fitting is determined, wavenumber is used as the independent variable for fitting calculation, absorbance is used as the dependent variable for fitting calculation, and least squares fitting is used to perform data fitting calculation on all baseline fitting sample points. Through calculation, a set of polynomial coefficients that can characterize the relationship between wavenumber and absorbance is obtained.

[0141] Based on the calculated set of polynomial coefficients, and following the polynomial function construction rules, a baseline drift function for the gas pipeline network is constructed. This baseline drift function uses wavenumber as an input parameter. After substituting the input wavenumber parameter into the function, the corresponding baseline absorbance value can be directly output.

[0142] The wavenumber sequence corresponding to the rough concentration value of the odorant in the gas pipeline network is retrieved. Each wavenumber value in the wavenumber sequence is substituted into the constructed baseline drift function in turn. The baseline absorbance offset corresponding to each wavenumber in the wavenumber sequence is obtained by calculating the baseline absorbance offset through the function, thus forming complete baseline absorbance offset data.

[0143] The system collects the current ambient temperature and pressure values ​​of the environment where the gas pipeline network is located, and simultaneously retrieves the system's preset reference temperature, reference pressure, temperature influence coefficient, pressure influence coefficient, and instrument calibration factor to complete the collection and organization of all parameters required for conversion.

[0144] The crude concentration values ​​of odorant in the gas pipeline network are subjected to deconvolution fitting. The concentration absorption peak corresponding to the crude concentration value of odorant is restored by fitting, and the peak shape characteristics and data distribution of the concentration absorption peak are fully presented.

[0145] The concentration absorption peak obtained by reduction is subjected to peak shape fitting processing. The peak center wavenumber of the concentration absorption peak is accurately extracted through fitting calculation. At the same time, the peak position weight coefficient that can characterize the peak position characteristics is calculated to obtain the core parameters of the peak shape.

[0146] By combining the collected parameters, peak center wavenumber, peak position weighting coefficient, and preset conversion rules, the numerical conversion is completed through a standardized calculation process to obtain the baseline concentration offset corresponding to the rough measured concentration value of the odorant in the gas pipeline network.

[0147] The instrument calibration factor is obtained in advance by the detection instrument used for monitoring the concentration of odorant in the gas pipeline network. The baseline absorbance offset is calculated by substituting the wavenumber sequence corresponding to the crude concentration value of the odorant into the baseline drift function. The peak weighting coefficient and peak center wavenumber are generated by performing deconvolution fitting on the concentration absorption peak obtained from the crude concentration value of the odorant, and then performing peak shape fitting on the concentration absorption peak. The current wavenumber position of the wavenumber sequence comes from the wavenumber sequence corresponding to the crude concentration value of the odorant. The peak shape attenuation coefficient and background constant are preset values ​​for monitoring the concentration of odorant in the gas pipeline network. The temperature influence coefficient and pressure influence coefficient are preset coefficients based on the monitoring environment characteristics of the gas pipeline network. The current ambient temperature value and current ambient pressure value are values ​​obtained by real-time detection of the monitoring environment of the gas pipeline network. The reference temperature and reference pressure are preset benchmark environmental values ​​for monitoring the concentration of odorant in the gas pipeline network. The natural constant is a fixed constant in the field of mathematics.

[0148] The calculation process quantifies and integrates various influencing factors related to the concentration of odorant in gas pipelines. Through multi-dimensional numerical calculations, the baseline absorbance shift generated by the original spectral response signal in the gas pipeline is converted into a baseline concentration shift that can directly reflect the deviation of the odorant concentration. This result is used to compare the difference with the rough concentration value of the odorant, and finally obtains the monitoring value of the trace concentration of odorant in the gas pipeline.

[0149] The difference between the current wavenumber position and the peak center wavenumber is squared and multiplied by the peak shape attenuation coefficient. The opposite of this value is then used as the exponent of the natural constant for exponential calculation. The result is multiplied by the peak position weighting coefficient and then by the background constant. The resulting value is then multiplied by the instrument correction factor and the baseline absorbance offset. The difference between the current ambient temperature and the reference temperature is multiplied by the temperature influence coefficient. The difference between the current ambient pressure and the reference pressure is multiplied by the pressure influence coefficient. The results of these two multiplications are added together and then multiplied by 1. The resulting value is then multiplied by the results of the previous multiplications. As the number of values ​​involved in the calculation increases, the final baseline concentration offset value increases, and vice versa.

[0150] The difference between the coarsely measured concentration of odorant in the gas pipeline network and the calculated baseline concentration offset is compared point by point. The baseline concentration offset at the corresponding position is subtracted from the coarsely measured concentration of odorant, and the calculated difference is used as the monitoring value of trace concentration of odorant in the gas pipeline network.

[0151] The beneficial effects are that this implementation process uses all discrete data points of the original spectral response signal as baseline fitting sample points, ensuring the sample integrity of the baseline drift function construction. Least square fitting with wavenumber and absorbance as variables allows the baseline drift function to accurately reflect the drift law of the spectral baseline. The baseline absorbance offset obtained by substituting the wavenumber sequence into the function has point-by-point matching accuracy. In the concentration offset conversion stage, environmental temperature and pressure parameters and instrument calibration factors are comprehensively collected. Combined with the core peak shape parameters extracted by deconvolution fitting and peak shape fitting, the calculation of baseline concentration offset fully fits the actual detection environment and spectral characteristics. Finally, by point-by-point subtraction of the coarse concentration value and the baseline concentration offset, accurate monitoring of odorant trace concentration is achieved. This effectively eliminates the influence of baseline drift, environmental factors and instrument errors on concentration monitoring, significantly improves the accuracy and reliability of odorant trace concentration monitoring results in gas pipeline networks, and provides core data support for the accurate control of odorant concentration in pipeline networks.

[0152] The signal warning module 105 is used to compare and analyze the monitoring value of the trace concentration of the odorant with a preset concentration safety threshold range to obtain an abnormal warning signal of the gas pipeline network.

[0153] In this embodiment of the invention, when the signal warning module performs a comparison and analysis of the odorant trace concentration monitoring value with a preset concentration safety threshold range to obtain an abnormal warning signal for the gas pipeline network, it is specifically used for:

[0154] Obtain historical monitoring values ​​of the trace concentration of the odorant, and construct a sliding window sequence for monitoring the concentration of the trace concentration of the odorant;

[0155] The concentration monitoring sliding window sequence is processed by moving average to obtain the window smoothed concentration value of the odorant trace concentration monitoring value;

[0156] The window smoothed concentration value is compared with a preset concentration safety threshold range. When the window smoothed concentration value is lower than the lower limit of the concentration safety threshold range, a low concentration anomaly marker is triggered, and the deviation between the window smoothed concentration value and the lower limit value is recorded.

[0157] When the window-smoothed concentration value is higher than the upper limit of the concentration safety threshold range, a high concentration anomaly marker is triggered, and the deviation between the window-smoothed concentration value and the upper limit value is recorded.

[0158] An abnormality warning signal for the gas pipeline network is generated based on the abnormality marker and the deviation amplitude.

[0159] All historical monitoring data of odorant trace concentration monitoring values ​​in the gas pipeline network are retrieved. The real-time odorant trace concentration monitoring values ​​and historical monitoring values ​​are integrated in chronological order of collection time. A sliding window range is defined according to a fixed amount of data. The concentration monitoring values ​​within the window are arranged in time series to construct a concentration monitoring sliding window sequence of odorant trace concentration monitoring values.

[0160] Moving average processing is performed on the constructed concentration monitoring sliding window sequence. The concentration data within the window are selected in chronological order for mean calculation. The mean value obtained from each calculation is sequentially mapped to the time node of the sliding window to complete the smoothing process of the entire concentration monitoring sliding window sequence, and the window smoothed concentration value of the odorant trace concentration monitoring value is obtained.

[0161] The system retrieves the preset safe threshold range for odorant concentration, and compares the obtained window smoothed concentration value with the upper and lower limits of the threshold range. If the window smoothed concentration value is lower than the lower limit of the safe threshold range, a low concentration anomaly flag is immediately triggered. At the same time, the difference between the window smoothed concentration value and the lower limit is calculated and recorded as the deviation amplitude.

[0162] If the value of the window smoothed concentration is higher than the upper limit of the concentration safety threshold range, a high concentration anomaly marker is immediately triggered. At the same time, the difference between the window smoothed concentration value and the upper limit value is calculated and recorded as the deviation amplitude.

[0163] By integrating the triggered anomaly marker types and recorded deviation amplitude values, and according to the system's preset warning signal generation rules, the anomaly markers and deviation amplitudes are embedded as core information into the designated data area of ​​the warning signal, thus completing the integration and encoding of information and generating an anomaly warning signal for the gas pipeline network that can be directly pushed.

[0164] The beneficial effects of this implementation process are that by integrating real-time and historical monitoring values ​​to construct a sliding window sequence for concentration monitoring, the concentration analysis has continuity and completeness in the time dimension. The moving average processing effectively eliminates the random fluctuation error of single monitoring data, allowing the smoothed concentration value of the window to truly reflect the actual change trend of the odorant concentration. Through precise comparison with the preset threshold range, it achieves accurate identification and marking of low and high concentration anomalies. At the same time, the recording of the deviation amplitude provides a quantitative reference for the degree of anomaly. Finally, the combination of anomaly marking and deviation amplitude generates an early warning signal, which contains both anomaly type information and quantitative data on the degree of anomaly. This achieves accurate and quantitative early warning of odorant concentration anomalies in gas pipeline networks, providing clear and effective data reference for pipeline network operation and maintenance personnel to take timely and targeted control measures, and significantly improving the timeliness and effectiveness of gas pipeline network concentration control.

[0165] The loss estimation module 106 is used to associate and store the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, so as to predict the odorant loss rate of the gas pipeline network.

[0166] In this embodiment of the invention, when the loss estimation module performs the process of associating and storing the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for:

[0167] Add a timestamp of the acquisition time to the odorant trace concentration monitoring value to obtain a time-stamped concentration data tuple of the odorant trace concentration monitoring value;

[0168] According to the time sequence, the time-stamped concentration data tuples are inserted into the circular storage buffer of the gas pipeline network to obtain the concentration time sequence data chain of the odorant trace concentration monitoring value;

[0169] The concentration time-series data chain is periodically scanned to obtain the current periodic concentration subsequence of the odorant trace concentration monitoring value;

[0170] Outlier removal is performed on the current periodic concentration subsequence to obtain a clean periodic concentration subsequence of the odorant trace concentration monitoring values;

[0171] By performing line fitting on the discrete data points in the clean periodic concentration subsequence, periodic concentration change segments of the odorant trace concentration monitoring values ​​are obtained.

[0172] By splicing the periodic concentration change segments together in chronological order, the odorant concentration change curve of the gas pipeline network is obtained, in order to predict the odorant consumption rate of the gas pipeline network.

[0173] When the loss estimation module performs the operation of splicing the periodic concentration change segments in chronological order to obtain the odorant concentration change curve of the gas pipeline network, in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for:

[0174] Identify the monotonically decreasing segment in the concentration change curve of the odorant, and extract the starting point concentration value, starting point time, ending point concentration value, and ending point time of the monotonically decreasing segment;

[0175] The concentration difference between the starting point concentration value and the ending point concentration value is calculated, as well as the time interval between the starting point time and the ending point time.

[0176] The instantaneous loss rate of the monotonically decreasing segment is determined based on the concentration difference and the time interval.

[0177] The instantaneous loss rate with the largest value is taken as the peak loss rate, and the instantaneous loss rate with the smallest value is taken as the trough loss rate.

[0178] The average of the peak loss rate, the trough loss rate, and the instantaneous loss rate is used as the predicted odorant loss rate of the gas pipeline network.

[0179] For each odorant trace concentration monitoring value in the gas pipeline network, match its collection time with the timestamp information, bind and integrate the timestamp with the corresponding concentration monitoring value to form a time-stamped concentration data tuple of odorant trace concentration monitoring value containing time information and concentration data.

[0180] According to the order of collection of timestamp records, all time-stamped concentration data tuples are sequentially inserted into the pre-set ring storage buffer of the gas pipeline network, forming a time-series data chain of odorant trace concentration monitoring values ​​arranged continuously in the time dimension in the buffer.

[0181] According to the fixed time period preset by the system, the concentration time series data chain in the circular storage buffer is fully scanned, all time-stamped concentration data tuples in that period are extracted, and integrated to form the current period concentration subsequence of the odorant trace concentration monitoring value.

[0182] The extracted current periodic concentration subsequence is screened, and abnormal concentration values ​​that deviate from the normal data range are identified through numerical distribution analysis. All abnormal concentration values ​​are removed from the subsequence, and values ​​that conform to the characteristics of normal data are retained to obtain a clean periodic concentration subsequence of odorant trace concentration monitoring values.

[0183] All discrete data points in the clean periodic concentration subsequence are sequentially connected according to the time stamp order. At the same time, the connected broken line is smoothed and fitted to form a periodic concentration change segment of the odorant trace concentration monitoring value that can reflect the periodic concentration change trend.

[0184] Following the chronological order of development, the periodic concentration change segments generated in each cycle are seamlessly spliced ​​together, ensuring continuous connection between the time and concentration dimensions of each segment, thus forming a complete odorant concentration change curve for the gas pipeline network. This provides data curve support for the subsequent prediction of odorant depletion rates.

[0185] Trend identification was performed on the spliced ​​odorant concentration change curve. The concentration change trend of the curve was analyzed segment by segment. The monotonically decreasing segment in the curve where the concentration continuously decreased over time was accurately located. At the same time, the starting point concentration value and starting time corresponding to the starting position of the monotonically decreasing segment, as well as the ending point concentration value and ending time corresponding to the ending position, were extracted.

[0186] The absolute difference between the extracted starting point concentration value and the ending point concentration value is calculated to obtain the concentration difference value of the monotonically decreasing segment. At the same time, the time length between the starting point time and the ending point time is calculated to obtain the time interval corresponding to the monotonically decreasing segment, thus completing the statistics of the two core data.

[0187] The concentration difference obtained from statistics is used as the numerator, and the corresponding time interval is used as the denominator. A division operation is performed, and the instantaneous loss rate of the odorant in the monotonically decreasing range is determined by the calculation result.

[0188] Summarize the instantaneous loss rates calculated for all monotonically decreasing segments, sort all values ​​by size, select the instantaneous loss rate with the largest value after sorting as the peak loss rate, and select the instantaneous loss rate with the smallest value after sorting as the trough loss rate.

[0189] The peak loss rate, trough loss rate, and the arithmetic mean of all instantaneous loss rates are integrated and calculated. The combined value of the three is taken as the final prediction result of the odorant loss rate of the gas pipeline network.

[0190] The beneficial effects of this implementation process are that by adding timestamps to concentration monitoring values ​​and constructing a time-series data chain, it achieves precise binding between concentration data and the time dimension, ensuring the temporal correlation of loss analysis. Regularly scanning and extracting periodic subsequences and removing outliers effectively filters invalid monitoring data, allowing clean periodic concentration subsequences to truly reflect the actual concentration changes in the pipeline network. The concentration change curve formed by connecting and fitting lines and splicing segments fully presents the overall trend of odorant concentration changes. Subsequently, by identifying monotonically decreasing segments and extracting key parameters to calculate the instantaneous loss rate, the calculation of the loss rate is made to fit the actual concentration consumption pattern of the pipeline network. Then, by comprehensively calculating the peak, trough, and average instantaneous loss rates, the prediction results are obtained, taking into account both the extreme characteristics and the overall average level of the loss rate. This makes the prediction results of odorant loss rate more comprehensive and accurate, providing a scientific and effective data reference for the replenishment planning and operation and maintenance scheduling of gas pipeline odorants, and improving the intelligence and refinement level of pipeline odorant management.

[0191] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0192] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0193] Finally, 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.

Claims

1. A gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines, characterized in that, The system includes a pressure decomposition module, a command generation module, a coarse concentration measurement module, a fine concentration measurement module, a signal early warning module, and a loss estimation module, wherein: The pressure decomposition module is used to perform empirical mode decomposition on the original pressure fluctuation time series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and use the instantaneous amplitude of the high-frequency intrinsic mode components as the pressure disturbance characteristic value of the gas pipeline network. The instruction generation module is used to query a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value, obtain the feedforward compensation coefficient of the gas pipeline network, and encode the feedforward compensation coefficient into the gas pressure compensation instruction of the gas pipeline network. The concentration coarse measurement module is used to receive the stable airflow generated after the gas pressure compensation command is executed, and based on the stable airflow, to identify the characteristic peaks of the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network. The concentration precision measurement module is used to perform polynomial fitting on the original spectral response signal to obtain the baseline drift function of the gas pipeline network, and to subtract the coarse concentration value of the odorant point by point according to the baseline drift function to obtain the trace concentration monitoring value of the odorant in the gas pipeline network, including; The discrete data points in the original spectral response signal are used as the baseline fitting sample points of the gas pipeline network. Set the order of the polynomial fitting, use wavenumber as the independent variable and absorbance as the dependent variable, and perform least squares fitting on the baseline fitting sample points to obtain the polynomial coefficient set. Based on the set of polynomial coefficients, a baseline drift function for the gas pipeline network is constructed. The baseline drift function takes wavenumber as input and outputs the corresponding baseline absorbance value. Obtain the wavenumber sequence corresponding to the crude concentration value of the odorant, and substitute the wavenumber sequence into the baseline drift function to obtain the baseline absorbance shift corresponding to the wavenumber sequence. According to the preset conversion rules, the baseline absorbance offset is converted into a baseline concentration offset; The difference between the crude concentration value of the odorant and the baseline concentration offset is compared to obtain the trace concentration monitoring value of the odorant in the gas pipeline network. When the concentration precision measurement module performs the conversion of the baseline absorbance offset into a baseline concentration offset according to a preset conversion rule, it is specifically used for: The current ambient temperature, current ambient pressure, reference temperature, reference pressure, temperature influence coefficient, pressure influence coefficient, and instrument correction factor of the gas pipeline network are obtained. The concentration absorption peak of the odorant is obtained by performing deconvolution fitting on the crude concentration value of the odorant. Peak shape fitting is performed on the concentration absorption peak to obtain the peak center wavenumber and peak position weighting coefficient of the concentration absorption peak; According to the preset conversion rules, the baseline concentration offset of the crude concentration value of the odorant is calculated; The formula for calculating the baseline concentration offset is: ； in, This indicates the baseline concentration offset. This represents the instrument's correction factor. This represents the baseline absorbance offset. Indicates the first The aforementioned peak position weighting coefficients. Indicates the first The peak center wavenumbers mentioned above This indicates the current wavenumber position of the wavenumber sequence. This represents the preset peak shape attenuation coefficient. This represents a preset background constant. This represents the temperature influence coefficient. This indicates the current ambient temperature value. This indicates the reference temperature. This represents the pressure influence coefficient. This indicates the current environmental pressure. This indicates the reference pressure. Represents the natural constant; The signal warning module is used to compare and analyze the monitoring value of the trace concentration of the odorant with a preset concentration safety threshold range to obtain an abnormal warning signal of the gas pipeline network. The loss estimation module is used to associate and store the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, so as to predict the odorant loss rate of the gas pipeline network.

2. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 1, characterized in that, When the gas pressure decomposition module performs empirical mode decomposition on the raw pressure fluctuation time-series data of the gas pipeline network to obtain the high-frequency intrinsic mode components of the gas pipeline network, and uses the instantaneous amplitude of the high-frequency intrinsic mode components as the gas pressure disturbance characteristic value of the gas pipeline network, it is specifically used for: The raw pressure fluctuation time series data of the node under test in the gas pipeline network is obtained within a continuous sampling period, and the raw pressure fluctuation time series data is filtered by a sliding window to obtain the smooth pressure time series data of the gas pipeline network. Local extremum points are detected in the smoothed pressure time series data to obtain the local maxima and local minima of the smoothed pressure time series data. By fitting the local maxima and local minima, the upper and lower envelopes of the gas pipeline network are obtained. The upper and lower envelopes are averaged to obtain the local mean curves of the gas pipeline network, and the local mean curves are removed to obtain the candidate intrinsic mode components of the gas pipeline network. The candidate intrinsic mode components are screened and a termination condition is determined to obtain the high-frequency intrinsic mode components of the gas pipeline network. The high-frequency intrinsic mode components are subjected to Hilbert transform to obtain the instantaneous frequency of the high-frequency intrinsic mode components, and the amplitude corresponding to the instantaneous frequency is used as the gas pressure disturbance characteristic value of the gas pipeline network.

3. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 1, characterized in that, When the instruction generation module executes the query of a preset disturbance-compensation mapping table based on the gas pressure disturbance characteristic value to obtain the feedforward compensation coefficient of the gas pipeline network, and encodes the feedforward compensation coefficient into a gas pressure compensation instruction for the gas pipeline network, it is specifically used for: Membership analysis is performed on the pressure disturbance characteristic values ​​to determine the target disturbance level range to which the pressure disturbance characteristic values ​​belong; Based on the target disturbance level range, the corresponding initial value of the feedforward compensation coefficient is extracted from the preset disturbance-compensation mapping table and used as the compensation coefficient to be corrected for the gas pipeline network. The difference between the pressure disturbance characteristic value and the interval endpoint value of the target disturbance level interval is compared to obtain the correction weight of the compensation coefficient to be corrected. Based on the correction weight, the compensation coefficient to be corrected is linearly adjusted to obtain the feedforward compensation coefficient of the air pressure disturbance characteristic value. The feedforward compensation coefficient is filled into the data field of the instruction frame in the gas pipeline network to obtain the gas pressure compensation instruction of the gas pipeline network.

4. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 1, characterized in that, When the concentration coarse measurement module executes the stable airflow generated after receiving the pressure compensation command, and based on the stable airflow, performs characteristic peak identification on the original spectral response signal of the gas pipeline network to obtain the coarse concentration value of the odorant in the gas pipeline network, it is specifically used for: By applying the pressure compensation command and monitoring the gas flow data of the gas pipeline network in real time, a stable gas flow of the gas pipeline network is obtained. The original spectral response signal of the gas pipeline network is parsed to extract a continuous spectral data frame sequence corresponding to the period of stable airflow generation. Noise estimation is performed on the continuous spectral data frame sequence to obtain the local baseline noise level of the original spectral response signal, and the peak detection threshold of the original spectral response signal is set according to the local baseline noise level. Based on the peak detection threshold, the candidate characteristic peaks of the original spectral response signal are obtained by traversing the continuous spectral data frame sequence. The peak wavelength position and peak height of the candidate characteristic peaks are extracted, and combined with the preset standard odorant characteristic spectral library, the effective characteristic peaks of the original spectral response signal are obtained; The effective characteristic peaks are quantitatively inverted to obtain the rough concentration value of the odorant in the gas pipeline network.

5. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 1, characterized in that, When the signal early warning module compares and analyzes the monitored value of the odorant trace concentration with a preset concentration safety threshold range to obtain an abnormal early warning signal for the gas pipeline network, it is specifically used for: Obtain historical monitoring values ​​of the trace concentration of the odorant, and construct a sliding window sequence for monitoring the concentration of the trace concentration of the odorant; The concentration monitoring sliding window sequence is processed by moving average to obtain the window smoothed concentration value of the odorant trace concentration monitoring value; The window smoothed concentration value is compared with a preset concentration safety threshold range. When the window smoothed concentration value is lower than the lower limit of the concentration safety threshold range, a low concentration anomaly marker is triggered, and the deviation between the window smoothed concentration value and the lower limit value is recorded. When the window-smoothed concentration value is higher than the upper limit of the concentration safety threshold range, a high concentration anomaly marker is triggered, and the deviation between the window-smoothed concentration value and the upper limit value is recorded. An abnormality warning signal for the gas pipeline network is generated based on the abnormality marker and the deviation amplitude.

6. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 1, characterized in that, When the loss estimation module associates and stores the odorant trace concentration monitoring value with the corresponding timestamp to obtain the odorant concentration change curve of the gas pipeline network, in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for: Add a timestamp of the acquisition time to the odorant trace concentration monitoring value to obtain a time-stamped concentration data tuple of the odorant trace concentration monitoring value; According to the time sequence, the time-stamped concentration data tuples are inserted into the circular storage buffer of the gas pipeline network to obtain the concentration time sequence data chain of the odorant trace concentration monitoring value; The concentration time-series data chain is periodically scanned to obtain the current periodic concentration subsequence of the odorant trace concentration monitoring value; Outlier removal is performed on the current periodic concentration subsequence to obtain a clean periodic concentration subsequence of the odorant trace concentration monitoring values; By performing line fitting on the discrete data points in the clean periodic concentration subsequence, periodic concentration change segments of the odorant trace concentration monitoring values ​​are obtained. By splicing the periodic concentration change segments together in chronological order, the odorant concentration change curve of the gas pipeline network is obtained, in order to predict the odorant consumption rate of the gas pipeline network.

7. The gas pressure stabilization and odorant trace concentration monitoring system for gas pipelines as described in claim 6, characterized in that, When the loss estimation module performs the operation of splicing the periodic concentration change segments in chronological order to obtain the odorant concentration change curve of the gas pipeline network, in order to predict the odorant loss rate of the gas pipeline network, it is specifically used for: Identify the monotonically decreasing segment in the concentration change curve of the odorant, and extract the starting point concentration value, starting point time, ending point concentration value, and ending point time of the monotonically decreasing segment; The concentration difference between the starting point concentration value and the ending point concentration value is calculated, as well as the time interval between the starting point time and the ending point time. The instantaneous loss rate of the monotonically decreasing segment is determined based on the concentration difference and the time interval. The instantaneous loss rate with the largest value is taken as the peak loss rate, and the instantaneous loss rate with the smallest value is taken as the trough loss rate. The average of the peak loss rate, the trough loss rate, and the instantaneous loss rate is used as the predicted odorant loss rate of the gas pipeline network.

Images