Safety Monitoring Method and System for Gas Pressure Regulators Based on Dual Time Windows

Abstract

This invention provides a safety monitoring method and system for gas pressure regulating devices based on a dual-time window, relating to the field of gas pressure regulating device safety monitoring technology. The method includes the following steps: real-time acquisition of the current pressure data stream of the gas pressure regulating device; continuous monitoring of the pressure data stream based on a first time window; when the pressure data stream continuously exceeds a first threshold within the first time window, a transient anomaly is determined to have occurred, the anomaly trigger time point is recorded, and two parallel processing branches are triggered simultaneously; based on the results of the two parallel processing branches, an early warning decision is executed. This method can not only capture transient anomalies with small amplitudes caused by diaphragm micro-damage in real time, but also detect trend anomalies caused by pressure continuously exceeding a high threshold.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring technology for gas pressure regulating devices, specifically to a safety monitoring method and system for gas pressure regulating devices based on dual time windows. Background Technology

[0002] Pressure regulating devices in gas transmission and distribution networks are core equipment for ensuring stable gas supply pressure and safe system operation. The health of the diaphragm, its core component, directly determines the pressure regulating performance. Diaphragm failure (such as fatigue, cracks, and perforation) is a progressive process from the accumulation of microscopic damage to macroscopic functional failure.

[0003] Currently, the industry commonly uses pressure over-limit alarm systems (such as SCADA systems) based on fixed safety thresholds for monitoring. This method has the following significant drawbacks: (1) The early warning is seriously delayed: the alarm is only triggered after the pressure value has exceeded the safety boundary and the fault has actually occurred, completely losing the window of preventive maintenance before the fault occurs, and the occurrence of accidents cannot be avoided; (2) "Blindness" to early failures: Unable to identify small, transient abnormal fluctuations in pressure amplitude that are still within the safe range in the early stages of diaphragm performance degradation (such as decreased elasticity or microcracks). These fluctuations are key signs of cumulative diaphragm damage. (3) The false alarm rate remains high: relying solely on single-point pressure data makes it impossible to effectively distinguish pressure changes caused by normal system disturbances such as upstream pressure fluctuations and downstream gas consumption changes from changes caused by diaphragm failures of the regulator itself, resulting in frequent false alarms and consuming operation and maintenance resources. (4) Lack of state quantification and prediction capabilities: It only provides a binary judgment of "normal / fault", and cannot quantify the degree of health degradation of the membrane, nor can it predict its remaining service life. The operation and maintenance decision can only rely on fixed maintenance cycles or passive response after a fault, resulting in a low level of intelligence.

[0004] Therefore, existing technologies cannot meet the urgent need for early, accurate, and quantitative early warning of progressive diaphragm failures in gas pressure regulating devices. Summary of the Invention

[0005] The purpose of this invention is to provide a safety monitoring method for gas pressure regulating devices based on dual time windows. Based on dual time windows of different lengths, two monitoring logics are executed on the pressure data, and pipeline propagation is introduced for cross-validation to verify the detected anomalies. This method can not only capture short-term anomalies with small amplitude caused by diaphragm micro-damage in real time, but also detect trend anomalies caused by pressure continuously exceeding a high threshold.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following solution: A safety monitoring method for gas pressure regulating devices based on dual time windows, the method comprising the following steps: Real-time acquisition of pressure data stream from the current gas pressure regulating device; Based on continuous monitoring of the pressure data stream within a first time window, when the pressure data stream continuously exceeds the first threshold within the first time window, a transient anomaly is determined, the anomaly trigger time is recorded, and the following two parallel processing branches are triggered simultaneously: First branch: Locate at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimate the impact time of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time, and analyze whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time. If so, mark the brief anomaly as a valid self-anomaly. The second branch: Based on the abnormal trigger time point, a second time window is started to monitor the pressure data stream and detect whether the pressure data stream continuously exceeds the second threshold within the second time window; the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold; Based on the results of the two parallel processing branches, an early warning decision is executed: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

[0007] Furthermore, in the first branch, the process of estimating the duration of the impact of a brief anomaly on other gas pressure regulating devices based on the anomaly trigger time is as follows: Obtain the pipeline distance and pipeline parameters between the current gas pressure regulating device and other gas pressure regulating devices; Based on real-time collected flow parameters, gas state parameters, and pipeline parameters, the propagation speed of pressure waves generated by transient anomalies in the pipeline is calculated. Based on the propagation speed and pipeline distance, the estimated time required for the pressure wave generated by the brief anomaly to propagate from the current gas pressure regulating device to other gas pressure regulating devices is calculated. The time period affected is defined as the time between the abnormal triggering point and the estimated time required.

[0008] Furthermore, in the first branch, the process of analyzing whether there is a propagation correlation between the pressure data stream and transient anomalies within the affected time period is as follows: Extract the first pressure change characteristic curve corresponding to the occurrence of a brief abnormality in the current gas pressure regulating device; Extract the second pressure change characteristic curve of other gas pressure regulating devices during the affected time period; A similarity analysis is performed on the first pressure change characteristic curve and the second pressure change characteristic curve. If the calculated similarity value exceeds a preset similarity threshold, it is determined that they have a propagation correlation.

[0009] Furthermore, in the second branch, starting the second time window based on the exception trigger time point means generating a second time window of fixed length with the exception trigger time point as the starting point of the second time window.

[0010] Furthermore, after starting a second time window to monitor the pressure data stream based on the abnormal trigger time point, if no pressure data stream is detected continuously exceeding the second threshold within the second time window, the second time window ends after reaching its fixed length; when the second time window is started again to monitor the pressure data stream from a new abnormal trigger time point, the starting point of the second time window jumps to the new abnormal trigger time point.

[0011] Furthermore, the process of generating a quantitative assessment result of the membrane health status based on the effective self-abnormality count is as follows: according to the preset correspondence between the count interval and the membrane health level, the current effective self-abnormality count is mapped to the corresponding health level, which is used as the quantitative assessment result.

[0012] Furthermore, the first warning information is a predictive warning based on the diaphragm's health status, used to assess cumulative diaphragm damage; the second warning information is an immediate warning based on the determination of current continuous pressure abnormality; the urgency of the fault indicated by the second warning information is higher than that of the first warning information.

[0013] A safety monitoring system for a gas pressure regulating device based on a dual-time window, employing the aforementioned safety monitoring method for a gas pressure regulating device based on a dual-time window, includes: The data acquisition module collects the pressure data stream of the current gas pressure regulator in real time. The first monitoring module continuously monitors the pressure data stream based on the first time window. When it detects that the pressure data stream continuously exceeds the first threshold within the first time window, it determines that a brief abnormality has occurred, records the time point of the abnormality trigger, and outputs the corresponding trigger signal. The anomaly verification module, in response to the trigger signal, locates at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimates the impact time period of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time point, and analyzes whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time period. If so, the brief anomaly is marked as a valid self-anomaly. The second monitoring module, in response to the trigger signal, starts a second time window to monitor the pressure data stream based on the abnormal trigger time point, and detects whether the pressure data stream continuously exceeds the second threshold within the second time window; the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold; The early warning decision module is connected to the anomaly verification module and the second monitoring module respectively: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

[0014] The beneficial effects of this invention are: This invention provides a safety monitoring method for gas pressure regulating devices based on dual time windows. It mainly involves setting a first time window and a second time window with different lengths and thresholds, executing two monitoring logics in parallel, and introducing a pipeline propagation cross-verification mechanism to identify the source of the detected anomalies.

[0015] Compared with the prior art, the present invention has the following significant advantages: (1) Early warning and process quantification of faults are realized: by capturing small and short-term initial anomalies through the first time window, and by statistically analyzing the frequency of their occurrence after verification, the invisible performance degradation process of the diaphragm is transformed into a quantifiable health indicator, thus realizing predictive warning before the occurrence of faults and greatly advancing the warning time. (2) Significantly improved the accuracy and reliability of early warning: Cross-verification through the physical characteristics of pipeline pressure propagation effectively distinguishes between equipment failure and normal disturbance of the pipeline system, fundamentally reducing the false alarm rate and making the early warning information more credible; (3) A dual early warning defense line covering the entire failure cycle has been constructed: the first time window monitoring combined with cumulative counting effectively warns of the slow and gradual "quantitative change" process; the second time window monitoring combined with high threshold directly captures the acute and sudden "qualitative change" failure. Moreover, the two paths work together to ensure that there are no blind spots in the full-cycle safety monitoring from early micro-damage to complete failure. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the steps of the safety monitoring method for the gas pressure regulating device in Embodiment 1 of the present invention.

[0017] Figure 2 This is a schematic diagram of a sudden emergency fault based on a dual time window in Embodiment 1 of the present invention.

[0018] Figure 3 This is a schematic diagram of a progressive micro-damage scenario of a membrane based on a dual time window in Embodiment 1 of the present invention. Detailed Implementation

[0019] 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 are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. 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.

[0020] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0021] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0022] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0023] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0024] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0025] The present invention will now be described in detail with reference to the accompanying drawings and embodiments: Example 1 In this embodiment, a safety monitoring method for a gas pressure regulating device based on a dual-time window is provided, such as... Figure 1 As shown, the method includes the following steps: Real-time acquisition of the outlet pressure data stream of the current gas pressure regulating device; The pressure data stream is continuously monitored based on the first time window. When the pressure data stream continuously exceeds the first threshold within the first time window, a brief anomaly is determined to have occurred. The time point of the anomaly is recorded, and two parallel processing branches are triggered simultaneously. Based on the results of the two parallel processing branches, an early warning decision is made.

[0026] In this embodiment, two parallel processing branches are triggered simultaneously, namely the first branch and the second branch.

[0027] Specifically, the first branch is the propagation verification branch, and its process is as follows: locate at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimate the impact time period of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time point, and analyze whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time period. If so, mark the brief anomaly as a valid self-anomaly.

[0028] In the first branch, the process of estimating the duration of the impact of a brief anomaly on other gas pressure regulating devices based on the anomaly trigger time is as follows: Obtain the pipeline distance and pipeline parameters between the current gas pressure regulating device and other gas pressure regulating devices; Based on real-time collected flow parameters, gas state parameters, and pipeline parameters, the propagation speed of pressure waves generated by transient anomalies in the pipeline is calculated. Based on the propagation speed and pipeline distance, the estimated time required for the pressure wave generated by the brief anomaly to propagate from the current gas pressure regulating device to other gas pressure regulating devices is calculated. The time period affected is defined as the time between the abnormal triggering point and the estimated time required.

[0029] In the first branch, the process of analyzing whether there is a propagation correlation between the pressure data stream and transient anomalies within the affected time period is as follows: Extract the first pressure change characteristic curve corresponding to the occurrence of a brief abnormality in the current gas pressure regulating device; Extract the second pressure change characteristic curve of other gas pressure regulating devices during the affected time period; A similarity analysis is performed on the first pressure change characteristic curve and the second pressure change characteristic curve. If the calculated similarity value exceeds a preset similarity threshold, it is determined that they have a propagation correlation.

[0030] Specifically, the second branch is the trend monitoring branch. The process is as follows: based on the abnormal trigger time point, the second time window is started to monitor the pressure data stream and detect whether the pressure data stream continuously exceeds the second threshold within the second time window; and the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold, that is, the first time window is a short time window and the second time window is a long time window.

[0031] The startup and monitoring logic of the second time window is applicable to the key of distinguishing fault types and realizing graded early warning. In the second branch, starting the second time window based on the abnormal trigger time point means taking the abnormal trigger time point as the starting point of the second time window and generating a second time window of fixed length.

[0032] like Figure 2 As shown, this illustrates the initial scenario of a sudden emergency failure (potentially caused by a diaphragm rupture, drive pressure failure, or other systemic problems). In this scenario, the anomaly trigger point T2 marks the end of the first time window [T1, T2], and the start of the second time window [T2, T3] marks the end of the first time window. The first threshold P1 of the first time window is less than the second threshold P2 of the second time window. It can be seen that during the subsequent monitoring period of the second time window, the pressure data stream continuously increases and rapidly exceeds the second threshold, forming a clear trend anomaly. This indicates that the initial brief anomaly is not an isolated event, but rather the initial stage or early characteristic of a larger, more persistent failure process (trend shift). This trend shift implies that the equipment may have suffered substantial functional damage requiring immediate intervention, necessitating an immediate second warning.

[0033] like Figure 3 As shown, this illustrates a typical scenario of progressive micro-damage to the diaphragm. In this case, the anomaly trigger point is the end of the first time window, and the start of the second time window is the end of the first time window. It can be seen that during the subsequent monitoring period of the second time window, the pressure data stream did not continue to increase or deviate for a long time, but quickly returned to the normal fluctuation range without reaching a higher second threshold. This indicates that the transient anomaly captured by the first time window is an independent, limited-amplitude fluctuation event, consistent with the characteristics of occasional regulatory misalignment caused by early micro-damage to the diaphragm. Furthermore, this fluctuation event, after being verified as a valid self-anomaly by the first branch, will be included in the statistics. However, in existing technologies, a large fixed time window (slow response) or a high fixed alarm threshold (insufficient sensitivity) is commonly used. Such minute initial anomalies cannot be effectively identified and recorded, thus missing the earliest warning opportunity.

[0034] At this point, the method of this invention will produce a completely different response: based on the monitoring results of the second branch, the system will immediately generate a second early warning message (an immediate emergency early warning), prompting maintenance personnel to take immediate action. Simultaneously, the system will reset the accumulated valid self-anomaly count in the first branch to zero, because this is no longer a gradual process requiring the accumulation of minor events for early warning, but rather an acute problem that has already erupted.

[0035] In summary, by setting a first time window and a second time window with different lengths and thresholds, and having them perform monitoring in parallel based on the same trigger point, a dual-dimensional analysis of the same abnormal event (the instantaneousness and persistence of amplitude / duration) is achieved. This design cleverly matches different modes of fault development: short windows and low thresholds are dedicated to capturing and quantifying the accumulation of "quantitative changes"; long windows and high thresholds are used to capture and confirm the occurrence of "qualitative changes." This is precisely the core reason why this invention can overcome the two major shortcomings of existing technologies: "insensitivity to minor early faults" and "lack of prediction of sudden serious faults."

[0036] Specifically, the system captures transient pressure anomalies caused by diaphragm micro-damage in real time through a short time window and low threshold. Once such anomalies are captured, two branches are immediately triggered in parallel: propagation verification and trend monitoring. The propagation verification branch analyzes pressure data from downstream related pressure regulating devices to confirm whether the anomaly originates from the target device itself. The trend monitoring branch, on the other hand, monitors for persistent pressure drift that indicates severe diaphragm damage through a long time window and high threshold. Finally, the system issues predictive warnings based on the cumulative frequency of verified valid anomalies, or immediate emergency warnings based on persistent trend anomalies.

[0037] Furthermore, in the actual operation of a gas pressure regulating device, a brief anomaly caused by micro-damage to the diaphragm typically involves a complete "generation-recovery" process in its pressure change waveform. Figure 3 As can be seen, when the pressure deviates from the normal value within the first time window, it gradually recovers. However, during this recovery process, the pressure curve continues to exceed the first threshold for a period of time. If left unaddressed, the first time window may misjudge this fluctuation during the recovery phase as a new, independent, and transient anomaly, leading to repeated counting of a single fault event, interfering with the statistical accuracy of the "effective self-anomaly count," and potentially causing false alarms.

[0038] To accurately count independent fault events and avoid misjudgments, this invention sets a silent period after detecting a brief anomaly. During this period, the system ignores fluctuations that may trigger the threshold again during pressure recovery, thereby preventing the recovery process of a single anomaly from being mistakenly counted as a new anomaly. This ensures the accuracy of effective self-anomaly counting and improves the overall reliability of the early warning system.

[0039] Based on the results of the two branches above, execute the early warning decision: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. The process of generating a quantitative assessment result of the membrane health status based on the effective self-abnormality count is as follows: according to the preset correspondence between the count interval and the membrane health level, the current effective self-abnormality count is mapped to the corresponding health level, which is used as the quantitative assessment result. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

[0040] The first warning information is a predictive warning based on the health status of the diaphragm, used to assess the cumulative damage to the diaphragm; the second warning information is an immediate warning based on the determination of the current continuous pressure abnormality; the urgency of the fault indicated by the second warning information is higher than that of the first warning information.

[0041] Based on the above principles, the present invention will be further explained and described as follows: In this embodiment, the current gas pressure regulating device is a key pressure regulator (denoted as R0) in a gas transmission and distribution pipeline network. Pressure data is collected in real time by a high-precision pressure sensor (sampling frequency of 10Hz) installed at its outlet, forming a pressure data stream P(t).

[0042] The length of the first time window, T1, is set to 10 seconds, and the first threshold ΔP1 is ±1.5% of the set pressure. The pressure data stream P(t) is continuously monitored, and the pressure value per second is compared with the set pressure value to obtain the relative deviation. When the relative deviation of 10 consecutive calculation points (i.e., within 10 seconds) exceeds ΔP1, a brief anomaly is determined to have occurred, and the precise time point at this time is recorded as the anomaly trigger time point t0.

[0043] Once a brief anomaly is detected and the anomaly trigger time t0 is recorded, both branches are started simultaneously.

[0044] Specifically, the process of starting the first branch is as follows: First, locate the other gas pressure regulating devices downstream of the current pressure regulating device R0. Based on the pre-stored pipeline topology diagram, determine that the other gas pressure regulating devices downstream of R0 are R1 and R2; Then, the estimated time period of impact of the transient anomaly on R1 and R2 is determined: First, the pipeline distances from R0 to R1 and R2 are obtained as L1 = 200 meters and L2 = 300 meters, respectively, with a pipeline diameter of DN200 for both. Based on the real-time collected flow data (current flow rate is 5000 m³ / h) and natural gas state parameters, the pressure wave propagation speed v is calculated to be approximately 300 m / s. Therefore, the propagation times to R1 and R2 are: t1 = L1 / v ≈ 0.67 seconds, t2 = L2 / v ≈ 1.0 seconds. Considering system errors, the impact time periods are set as [t0+t1-0.5, t0+t1+0.5] and [t0+t2-0.5, t0+t2+0.5], respectively, representing windows of 1 second in length. Next, the pressure data of R1 and R2 during the corresponding time period are extracted, and their pressure change characteristic curves are calculated respectively. At the same time, the pressure change characteristic curve of R0 in the [t0-1, t0+1] seconds (i.e. 1 second before and after the anomaly occurs) is extracted. Finally, a similarity analysis is performed: the dynamic time warping (DTW) algorithm is used to calculate the similarity distance between the pressure curves of R0 and R1, and R0 and R2. The similarity threshold is set to 0.8 (normalized similarity, 1 is completely identical). If the similarity between R0 and the curve of any other gas pressure regulator exceeds 0.8, the transient anomaly is determined to have propagation correlation and is marked as a valid self-anomaly.

[0045] Specifically, the process of starting the second branch is as follows: First, based on the abnormal trigger time point t0, a second time window is initiated. The length of the second time window, T2, is set to 5 minutes, and the second threshold ΔP2 is ±3.0% of the set pressure. Then, starting from t0, the pressure data stream is monitored for the next 5 minutes: the average pressure per minute is calculated, and it is checked whether the relative pressure deviation at 5 consecutive points (i.e., within 5 minutes) exceeds ΔP2. If so, an abnormal trend is determined.

[0046] Execute early warning decisions based on the results of both branches: For the first branch, the valid self-abnormality counter C is incremented by 1 for each valid self-abnormality detected. In this embodiment, the preset threshold N is 10 times. When the counter C reaches 10, a quantitative assessment result of the diaphragm health status is generated. Specifically, the health level is divided according to the value of the count C: C < 5 is healthy, 5 ≤ C < 10 is alert, and C ≥ 10 is a warning. At this time, the system generates a first warning message, which reads: "Regulator R0 diaphragm health warning: 10 valid abnormalities have been detected recently. It is recommended to arrange inspection and maintenance."

[0047] For the second branch, if the pressure is detected to continuously exceed ΔP2 within the 5-minute second time window, a second warning message is immediately generated: "Pressure regulator R0 has experienced a continuous pressure abnormality; please handle it immediately!" Simultaneously, the effective self-abnormality counter C in the first branch is reset to zero. If no trend abnormality is detected within the second time window, the branch ends after the 5-minute window closes, awaiting the next trigger.

[0048] In addition, after the second warning information is issued by the second branch, the system will interrupt the continuous monitoring of the first time window (i.e., enter a silent period). The silent period is set to 30 minutes to prevent the same continuous abnormal event from causing the first time window to be repeatedly triggered.

[0049] Through the above steps, early warning and immediate alarm for diaphragm failure of gas pressure regulating device are achieved, effectively distinguishing between its own failure and external interference, and providing quantitative assessment, thus providing a reliable basis for operation and maintenance decisions.

[0050] Example 2 A safety monitoring system for a gas pressure regulating device based on a dual-time window, employing the aforementioned safety monitoring method for a gas pressure regulating device based on a dual-time window, includes: The data acquisition module collects the pressure data stream of the current gas pressure regulator in real time. The first monitoring module continuously monitors the pressure data stream based on the first time window. When it detects that the pressure data stream continuously exceeds the first threshold within the first time window, it determines that a brief abnormality has occurred, records the time point of the abnormality trigger, and outputs the corresponding trigger signal. The anomaly verification module, in response to the trigger signal, locates at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimates the impact time period of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time point, and analyzes whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time period. If so, the brief anomaly is marked as a valid self-anomaly. The second monitoring module, in response to the trigger signal, starts a second time window to monitor the pressure data stream based on the abnormal trigger time point, and detects whether the pressure data stream continuously exceeds the second threshold within the second time window; the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold; The early warning decision module is connected to the anomaly verification module and the second monitoring module respectively: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A safety monitoring method for a gas pressure regulating device based on a dual-time window, characterized in that, The method includes the following steps: Real-time acquisition of pressure data stream from the current gas pressure regulating device; Based on continuous monitoring of the pressure data stream within a first time window, when the pressure data stream continuously exceeds the first threshold within the first time window, a transient anomaly is determined, the anomaly trigger time is recorded, and the following two parallel processing branches are triggered simultaneously: First branch: Locate at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimate the impact time of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time, and analyze whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time. If so, mark the brief anomaly as a valid self-anomaly. The second branch: Based on the abnormal trigger time point, a second time window is started to monitor the pressure data stream and detect whether the pressure data stream continuously exceeds the second threshold within the second time window; the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold; Based on the results of the two parallel processing branches, an early warning decision is executed: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

2. The safety monitoring method for a gas pressure regulating device based on a dual time window according to claim 1, characterized in that, In the first branch, the process of estimating the duration of the impact of a brief anomaly on other gas pressure regulating devices based on the anomaly trigger time is as follows: Obtain the pipeline distance and pipeline parameters between the current gas pressure regulating device and other gas pressure regulating devices; Based on real-time collected flow parameters, gas state parameters, and pipeline parameters, the propagation speed of pressure waves generated by transient anomalies in the pipeline is calculated. Based on the propagation speed and pipeline distance, the estimated time required for the pressure wave generated by the brief anomaly to propagate from the current gas pressure regulating device to other gas pressure regulating devices is calculated. The time period affected is defined as the time between the abnormal triggering point and the estimated time required.

3. The safety monitoring method for a gas pressure regulating device based on a dual time window according to claim 1, characterized in that, In the first branch, the process of analyzing whether there is a propagation correlation between the pressure data stream and transient anomalies within the affected time period is as follows: Extract the first pressure change characteristic curve corresponding to the occurrence of a brief abnormality in the current gas pressure regulating device; Extract the second pressure change characteristic curve of other gas pressure regulating devices during the affected time period; A similarity analysis is performed on the first pressure change characteristic curve and the second pressure change characteristic curve. If the calculated similarity value exceeds a preset similarity threshold, it is determined that they have a propagation correlation.

4. The safety monitoring method for a gas pressure regulating device based on a dual time window according to claim 1, characterized in that, In the second branch, starting the second time window based on the exception trigger time point means using the exception trigger time point as the starting point of the second time window and generating a second time window of fixed length.

5. A safety monitoring method for a gas pressure regulating device based on a dual-time window according to claim 4, characterized in that, After the second time window is started based on the abnormal trigger time point to monitor the pressure data stream, if the pressure data stream does not continuously exceed the second threshold within the second time window, the second time window will end after reaching its fixed length. When the second time window is started again to monitor the pressure data stream at a new anomaly trigger time point, the starting point of the second time window jumps to the new anomaly trigger time point.

6. The safety monitoring method for a gas pressure regulating device based on a dual time window according to claim 1, characterized in that, The process of generating a quantitative assessment result of the membrane health status based on the effective self-abnormality count is as follows: according to the preset correspondence between the count interval and the membrane health level, the current effective self-abnormality count is mapped to the corresponding health level, which is used as the quantitative assessment result.

7. The safety monitoring method for a gas pressure regulating device based on a dual time window according to claim 1, characterized in that, The first warning information is a predictive warning based on the health status of the membrane, used to assess the cumulative damage to the membrane; The second early warning information is an immediate warning based on the current determination of abnormal continuous pressure; The second warning message indicates a higher level of urgency than the first warning message.

8. A safety monitoring system for a gas pressure regulating device based on a dual-time window, characterized in that, The method for safety monitoring of a gas pressure regulating device based on a dual time window, as described in any one of claims 1-7, includes: The data acquisition module collects the pressure data stream of the current gas pressure regulator in real time. The first monitoring module continuously monitors the pressure data stream based on the first time window. When it detects that the pressure data stream continuously exceeds the first threshold within the first time window, it determines that a brief abnormality has occurred, records the time point of the abnormality trigger, and outputs the corresponding trigger signal. The anomaly verification module, in response to the trigger signal, locates at least one other gas pressure regulating device downstream of the current gas pressure regulating device, estimates the impact time period of the brief anomaly reaching the other gas pressure regulating device based on the anomaly trigger time point, and analyzes whether there is a propagation correlation between the pressure data stream and the brief anomaly during the impact time period. If so, the brief anomaly is marked as a valid self-anomaly. The second monitoring module, in response to the trigger signal, starts a second time window to monitor the pressure data stream based on the abnormal trigger time point, and detects whether the pressure data stream continuously exceeds the second threshold within the second time window; the length of the first time window is less than the length of the second time window, and the first threshold is less than the second threshold; The early warning decision module is connected to the anomaly verification module and the second monitoring module respectively: If the cumulative effective self-abnormal count in the first branch reaches the preset threshold, a quantitative assessment result of the membrane health status is generated based on the effective self-abnormal count, and a first warning message is issued. If the pressure data stream continuously exceeds the second threshold within the second time window, a second warning message is issued in the second branch, and the accumulated valid self-abnormal count in the first branch is cleared to zero.

Images