A pipeline leakage monitoring method and system based on infrasound

Abstract

The present invention relates to the technical field of pipeline monitoring, and discloses a pipeline leakage monitoring method and system based on infrasound. The pipeline leakage monitoring method comprises the following steps: S1: setting an infrasound sensor, determining the severity score of the leakage and the location of the leakage point; S2: obtaining historical alarm records, calculating a screening value, and determining whether the sub-pipeline is the first pipeline based on the screening value; S3: determining a target line based on the severity score on the first pipeline; calculating the omission ratio and adjusting the target line until the omission ratio meets the conditions; S4: determining a working sensor, and the infrasound sensors on the first pipeline other than the working sensor enter a dormant state. The present invention reduces the energy consumption of the infrasound sensor when monitoring the pipeline.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline monitoring, and in particular to a pipeline leakage monitoring method and system based on infrasound waves. Background Art

[0002] Due to the propagation characteristics of infrasound, such as minimal energy attenuation during propagation and strong penetration, it has potential in many applications, one of which is gas pipeline leakage monitoring. When a gas pipeline leaks, it is usually accompanied by rapid gas flow, which will trigger the generation of infrasound. Through specialized infrasound sensors, these low-frequency fluctuations can be detected, thereby promptly discovering abnormal conditions in the pipeline system.

[0003] Although infrasonic sensors have important applications in pipeline leakage monitoring, their long-term operation will lead to high energy consumption. In actual application scenarios, since infrasonic sensors need to continuously monitor low-frequency fluctuations in the pipeline, this continuous working mode often consumes a lot of power resources. Especially in large-scale pipeline systems, the number of sensors and the length of monitoring time further aggravate energy consumption. Therefore, how to reduce the energy consumption required for pipeline leakage monitoring while ensuring the monitoring quality has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a pipeline leakage monitoring method and system based on infrasound to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A pipeline leakage monitoring method based on infrasound waves comprises the following steps:

[0007] S1: Infrasonic sensors are arranged at equal intervals on the pipeline, and infrasonic signals inside the pipeline are collected by the infrasonic sensors. The infrasonic signals are preprocessed to extract characteristic parameters of the preprocessed infrasonic signals, where the characteristic parameters include frequency characteristics and energy distribution. The characteristic parameters are input into a pre-trained scoring estimation model, and the location of the leakage point is determined based on a time difference positioning algorithm, and the severity score of each leakage point on the pipeline is output;

[0008] S2: Marking a portion between two adjacent preset connection nodes in the pipeline as a sub-pipeline, obtaining a historical alarm record of the sub-pipeline, wherein the historical alarm record stores leakage alarm information within a preset period, wherein a single leakage alarm information includes a leakage point location and a corresponding severity score, calculating a screening value based on the historical alarm record, and determining whether the sub-pipeline is the first pipeline based on the screening value;

[0009] S3: sorting the severity scores of the leakage points when leakage occurs on the first pipeline in descending order to obtain a first order, obtaining the first leakage point in the first order, obtaining the distance parameters between the infrasonic wave sensors and the first leakage point, sorting the distance parameters in ascending order to obtain a second order, determining the corresponding infrasonic wave sensors based on the first three distance parameters in the second order, marking them as target sensors, and determining the target line based on the connecting line of the positions of the three target sensors;

[0010] The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold.

[0011] S4: acquiring the target line A and the number of omission ratio calculations at this time, determining the working sensor based on the target line A and the number of omission ratio calculations a, and controlling the infrasonic wave sensors on the first pipeline except the working sensor to enter a dormant state.

[0012] As a further solution of the present invention: in step S2, the process of determining the first pipeline includes:

[0013] Calculate filter values , represents the filter value used to determine the first pipeline, Indicates the sub-pipeline The severity rating of the leak when it occurs, Indicates the total number of leaks on the sub-pipeline. is the preset correction factor, For the length of the sub-pipeline, the sub-pipeline whose screening value is less than a preset screening value threshold is marked as the first pipeline.

[0014] As a further solution of the present invention: in step S1, the process of pre-training the score estimation model includes:

[0015] Establishing a database, wherein the database stores characteristic parameters with annotated severity scores;

[0016] A rating estimation model is established based on the machine learning model, and the rating estimation model is trained and verified based on the database, wherein the rating estimation model is trained using a back propagation algorithm, and the rating estimation model is verified using a five-fold cross validation.

[0017] As a further solution of the present invention: the characteristic parameters stored in the database are based on manually annotated severity scores.

[0018] As a further solution of the present invention: in the step S1, the process of preprocessing the infrasound signal includes denoising, filtering and enhancement.

[0019] As a further solution of the present invention: in step S4, the process of determining the working sensor includes:

[0020] Connecting the leakage point positions corresponding to the omission ratio calculation times by straight lines to obtain a leakage line segment, obtaining the starting point and the end point of the leakage line segment, wherein the severity score at the first position closest to the starting point is smaller than the severity score at the second position closest to the end point;

[0021] Establish a number axis, the origin of the number axis is the starting point, and the positive direction of the number axis points from the starting point to the end point, and the unit distance of the number axis is a preset value, generate coordinate points ( , ), Indicates that the distance between the number axis and the origin is The severity score when a leak occurs at the location of the =0;

[0022] Calculate theoretical position , Indicates the total number of times leakage occurs on the portion of the target pipeline corresponding to the target line A, determines the infrasonic sensor closest to the theoretical position, uses it as the key sensor, connects the position of the key sensor and the first position to obtain a first sub-segment, and connects the position of the key sensor and the second position to obtain a second sub-segment;

[0023] The first sub-line segment and the second sub-line segment are respectively used as leakage line segments, and corresponding working sensors are determined.

[0024] As a further solution of the present invention: the step S4 further includes the following steps:

[0025] When the duration of any acoustic wave sensor entering the dormant state reaches a preset duration threshold, a new working sensor is re-determined.

[0026] A pipeline leakage monitoring system based on infrasound, characterized by comprising:

[0027] An initial module is used to set infrasonic sensors at equal intervals on the pipeline system, collect infrasonic signals inside the pipeline based on the infrasonic sensors, preprocess the infrasonic signals, extract characteristic parameters of the preprocessed infrasonic signals, the characteristic parameters include frequency characteristics and energy distribution, input the characteristic parameters into a pre-trained scoring estimation model, determine the location of the leakage point based on the time difference positioning algorithm, output the severity score of each leakage point on the pipeline, and determine the location of the leakage point based on the time difference positioning algorithm;

[0028] A screening module, used for marking a pipeline portion between two adjacent preset connection nodes in the pipeline system as a sub-pipeline, obtaining a historical alarm record of the sub-pipeline, wherein the historical alarm record stores leakage alarm information within a preset period, wherein a single leakage alarm information includes a leakage point location and a corresponding severity score, calculating a screening value based on the historical alarm record, and determining whether the sub-pipeline is the first pipeline based on the screening value;

[0029] an optimization module, configured to sort the severity scores of each leakage point when leakage occurs on the first pipeline in descending order to obtain a first order, obtain the first leakage point at the first position in the first order, obtain the distance parameters between each of the infrasonic wave sensors and the first leakage point, sort each of the distance parameters in ascending order to obtain a second order, determine the corresponding infrasonic wave sensor based on the first three distance parameters in the second order, mark it as a target sensor, and determine a target line based on a line connecting the positions of the three target sensors;

[0030] The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold.

[0031] The adjustment module is used to obtain the target line A and the number of calculations a for calculating the missing ratio at this time, determine the working sensor based on the target line A and the number of calculations a for the missing ratio, and control the infrasonic wave sensors on the first pipeline except the working sensor to enter a dormant state.

[0032] Beneficial effects of the present invention: In this scheme, firstly, a method for determining the severity score and the location of the leakage point is given; based on the infrasonic sensor, the infrasonic signal inside the pipeline is collected and preprocessed, and the key characteristic parameters such as frequency characteristics and energy distribution are extracted, and the severity score of the leakage and the location of the leakage point are output through the pre-trained scoring estimation model, so as to accurately locate the leakage point and evaluate the severity of the leakage; then, by analyzing the historical alarm records and judging whether to mark a part of the pipeline as the first pipeline based on the screening value, the pipeline system is reasonably partitioned and prioritized; by calculating the screening value, those pipeline sections with frequent and serious leakage can be identified, and then they are monitored in key areas, providing a more accurate basis for the selection of working sensors for the subsequent work, helping to reduce energy consumption without sacrificing the monitoring effect; then, by sorting the severity of the leakage and the distance of the leakage point from the infrasonic sensor, it is possible to accurately determine which sensors are the shortest distance from the leakage point, and then determine the target sensor In the process of selecting the target sensor, by considering the location of the leak and the corresponding severity of the leak, the configuration of the target sensor can be repeatedly adjusted to optimize the leak monitoring area. The benefit of this step is that by focusing the monitoring on the most critical pipeline section, invalid or redundant monitoring is reduced, unnecessary sensor work is reduced, and energy consumption is further reduced. At the same time, it ensures that the monitoring of key leaks is not missed, and the monitoring accuracy is ensured; finally, by analyzing the arrangement pattern of the leak points and combining the severity scores of historical leaks, the selection and adjustment of the working sensors are further optimized; according to the different severities at both ends of the leak line segment, the position is optimized through the number axis, and finally the most suitable working sensor can be selected, thereby realizing intelligent sensor working state adjustment, avoiding the continuous activation of all sensors, but only activating the most relevant sensors according to needs; avoiding the continuous operation of the sensor, greatly reducing the system energy consumption, and ensuring that the monitoring of important leaks is not affected, ensuring the accuracy and efficiency of monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below in conjunction with the accompanying drawings.

[0034] Figure 1 The present invention is a schematic flow chart of a pipeline leakage monitoring method based on infrasound waves. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] See also Figure 1 As shown, the present invention is a pipeline leakage monitoring method based on infrasound, comprising the following steps:

[0037] S1: Infrasonic sensors are arranged at equal intervals on the pipeline, and infrasonic signals inside the pipeline are collected by the infrasonic sensors. The infrasonic signals are preprocessed to extract characteristic parameters of the preprocessed infrasonic signals, where the characteristic parameters include frequency characteristics and energy distribution. The characteristic parameters are input into a pre-trained scoring estimation model, and the location of the leakage point is determined based on a time difference positioning algorithm, and the severity score of each leakage point on the pipeline is output;

[0038] S2: Marking a portion of the pipeline between two adjacent preset connection nodes in the pipeline as a sub-pipeline, obtaining a historical alarm record of the sub-pipeline, wherein the historical alarm record stores leakage alarm information within a preset period, wherein each leakage alarm information includes a leakage point location and a corresponding severity score, calculating a screening value based on the historical alarm record, and determining whether the sub-pipeline is the first pipeline based on the screening value;

[0039] S3: sorting the severity scores of the leakage points when leakage occurs on the first pipeline in descending order to obtain a first order, obtaining the first leakage point D in the first order, obtaining the distance parameters between each of the infrasonic wave sensors and the first leakage point D, sorting each of the distance parameters in ascending order to obtain a second order, determining the corresponding infrasonic wave sensors based on the first three distance parameters in the second order, marking them as target sensors, and determining the target line based on the connecting line of the positions of the three target sensors;

[0040] The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where B represents the omission ratio, M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold.

[0041] S4: obtaining the target line A and the number of calculations a for calculating the missing ratio at this time, determining the working sensor based on the target line A and the number of calculations a for the missing ratio, and controlling the infrasonic wave sensors on the first pipeline except the working sensor to enter a dormant state.

[0042] It should be noted that, in the embodiment of the present application, a method for determining the severity score and the location of the leakage point is first given; based on the infrasonic sensor, the infrasonic signal inside the pipeline is collected and preprocessed, key characteristic parameters such as frequency characteristics and energy distribution are extracted, and the severity score of the leakage and the location of the leakage point are output through the pre-trained scoring estimation model, so as to accurately locate the leakage point and evaluate the severity of the leakage; thereafter, by analyzing the historical alarm records and judging whether to mark a certain part of the pipeline as the first pipeline based on the screening value, the pipeline system is reasonably partitioned and prioritized; by calculating the screening value, those pipeline sections with frequent and serious leakage can be identified, and then they can be monitored in key areas, providing a more accurate basis for the selection of working sensors for the subsequent work, helping to reduce energy consumption without sacrificing the monitoring effect; then, by sorting the severity of the leakage and the distance of the leakage point from the infrasonic sensor, it can be Accurately determine which sensors are the shortest distance from the leak point, and then determine the location of the target sensor. In the process of selecting the target sensor, by considering the location of the leak and the corresponding leak severity, the configuration of the target sensor can be repeatedly adjusted to optimize the leak monitoring area. The benefit of this step is that by focusing the monitoring on the most critical pipeline section, invalid or redundant monitoring is reduced, unnecessary sensor work is reduced, and energy consumption is further reduced. At the same time, it ensures that the monitoring of key leaks is not missed and the accuracy of monitoring is ensured; finally, by analyzing the arrangement pattern of the leak points and combining the severity scores of historical leak points, the selection and adjustment of the working sensors are further optimized; according to the different severities at both ends of the leak line segment, the position is optimized through the number axis, and finally the most suitable working sensor can be selected, thereby realizing intelligent sensor working state adjustment, avoiding the continuous activation of all sensors, and activating only the most relevant sensors as needed.

[0043] It should be noted that the connection node includes but is not limited to the location where two pipes are connected, such as the location of devices such as tees, and the location where the pipe turns (i.e. the location where the pipe is not on a straight line at the beginning).

[0044] In another preferred embodiment of the present invention, in step S2, the process of determining the first pipeline includes: calculating the screening value , represents the filter value used to determine the first pipeline, Indicates the sub-pipeline The severity rating of the leak when it occurs, Indicates the total number of leaks on the sub-pipeline. is the preset correction factor, For the length of the sub-pipeline, the sub-pipeline whose screening value is less than a preset screening value threshold is marked as the first pipeline.

[0045] In another preferred embodiment of the present invention, in step S1, the process of pre-training the score estimation model includes:

[0046] Establishing a database, wherein the database stores characteristic parameters with annotated severity scores;

[0047] A rating estimation model is established based on the machine learning model, and the rating estimation model is trained and verified based on the database, wherein the rating estimation model is trained using a back propagation algorithm, and the rating estimation model is verified using a five-fold cross validation.

[0048] It is worth noting that by establishing a database and storing the characteristic parameters of the labeled severity scores, a score estimation model is established based on the machine learning model, and trained through the back propagation algorithm and verified through five-fold cross validation, so as to ensure the accuracy and robustness of the score estimation model; through precise data training, the score estimation model can more accurately predict the severity of the leak and the location of the leak point, thereby improving the reliability and accuracy of monitoring; the parameters of the model are optimized through the back propagation algorithm to ensure that the model can adapt to different leakage characteristics in the pipeline system, and the stability of the model under different conditions is ensured through cross validation.

[0049] In another preferred embodiment of the present invention, the characteristic parameters stored in the database are based on manually annotated severity scores.

[0050] In another preferred embodiment of the present invention, in step S1, the process of preprocessing the infrasound wave signal includes denoising, filtering and enhancement.

[0051] It can be understood that the denoising process refers to removing useless signals introduced into the signal due to environmental noise or other external interference factors, ensuring that only real infrasonic signals can be analyzed in the subsequent processing process. Denoising can use the wavelet transform method to remove noise in different frequency bands through multi-scale analysis; filtering is to remove frequency components beyond the frequency band of interest through filtering in a specific frequency band, so that the signal is more focused on the frequency range of interest. Filtering can use a bandpass filter to filter out unnecessary high-frequency or low-frequency noise by setting a frequency range; enhancement is to strengthen specific features in the signal through an algorithm, so that the effective signal that may have been weaker is more prominent, which is convenient for subsequent analysis and modeling. Enhancement can use an adaptive enhancement method to adjust the amplitude of the signal through adaptive gain to make important leakage signal features more prominent.

[0052] In another preferred embodiment of the present invention, in step S4, the process of determining the working sensor includes:

[0053] Connect the leakage point positions corresponding to the omission ratio calculation times a by straight lines to obtain a leakage line segment, obtain the starting point f1 and the end point f2 of the leakage line segment, and the severity score at the first position F1 closest to the starting point f1 is smaller than the severity score at the second position F2 closest to the end point f2;

[0054] Establish a number axis, the origin of which is the starting point f1, and the positive direction of the number axis points from the starting point f1 to the end point f2, and the unit distance of the number axis is a preset value, and generate coordinate points ( , ), Indicates that the distance between the number axis and the origin is The severity score when a leak occurs at the location of the =0;

[0055] Calculate theoretical position , Indicates the total number of times leakage occurs on the portion of the target pipeline corresponding to the target line A, determines the infrasonic sensor closest to the theoretical position, uses it as the key sensor G1, connects the position of the key sensor G1 and the first position F1 to obtain a first sub-segment GF1, and connects the position of the key sensor G1 and the second position F2 to obtain a second sub-segment GF2;

[0056] The first sub-segment GF1 and the second sub-segment GF2 are respectively used as leakage segments, and corresponding working sensors are determined.

[0057] It can be understood that the theoretical position is the position of the theoretical center of gravity on the number axis obtained by using the severity score as the weight.

[0058] In another preferred embodiment of the present invention, the step S4 further includes the following steps: when the duration of any acoustic wave sensor entering the dormant state reaches a preset duration threshold, a new working sensor is re-determined.

[0059] By monitoring the working status and sleep time of sensors, potential performance degradation or failure of certain sensors due to long-term inactivity can be avoided. At the same time, working sensors can be dynamically adjusted according to real-time pipeline leakage monitoring needs to ensure that the monitoring coverage is not affected and to increase the service life of the sensors.

[0060] A pipeline leakage monitoring system based on infrasound waves, comprising:

[0061] Initial module: used to set infrasonic sensors at equal intervals on the pipeline system, collect infrasonic signals inside the pipeline based on the infrasonic sensors, pre-process the infrasonic signals, extract characteristic parameters of the pre-processed infrasonic signals, the characteristic parameters include frequency characteristics and energy distribution, input the characteristic parameters into the pre-trained scoring estimation model, determine the location of the leakage point based on the time difference positioning algorithm, output the severity score of each leakage point on the pipeline, and determine the location of the leakage point based on the time difference positioning algorithm;

[0062] Screening module: used for marking the pipeline part between two adjacent preset connection nodes in the pipeline system as a sub-pipeline, obtaining the historical alarm record of the sub-pipeline, wherein the historical alarm record stores leakage alarm information within a preset period, wherein each leakage alarm information includes a leakage point location and a corresponding severity score, calculating a screening value based on the historical alarm record, and determining whether the sub-pipeline is the first pipeline based on the screening value;

[0063] Optimization module: used for sorting the severity scores of each leakage point when leakage occurs on the first pipeline in descending order to obtain a first order, obtaining the first leakage point in the first order, obtaining the distance parameters between each of the infrasonic wave sensors and the first leakage point, sorting each of the distance parameters in ascending order to obtain a second order, determining the corresponding infrasonic wave sensor based on the first three distance parameters in the second order, marking it as a target sensor, and determining the target line based on the connecting line of the positions of the three target sensors;

[0064] The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold.

[0065] Adjustment module: used to obtain the target line A and the number of calculations a for calculating the missing ratio at this time, determine the working sensor based on the target line A and the number of calculations a for the missing ratio, and control the infrasonic wave sensors on the first pipeline except the working sensor to enter a dormant state.

[0066] The technical solution of the embodiment of the present application first outputs the severity score of the leak and the location of the leak point through a pre-trained scoring estimation model; then, by analyzing the historical alarm records, the pipeline system is reasonably partitioned and prioritized; by calculating the screening value, those pipeline sections with frequent and serious leaks can be identified, and then they are monitored intensively, providing a more accurate basis for selecting working sensors for the next work, helping to reduce energy consumption without sacrificing monitoring effects; then, the position of the target sensor is determined and repeatedly adjusted to optimize the leak point monitoring area. The benefit of this step is that by focusing the monitoring on the most critical pipeline section, invalid or redundant monitoring is reduced, unnecessary sensor work is reduced, and energy consumption is further reduced, while ensuring that the monitoring of key leak points is not missed, ensuring monitoring accuracy; finally, the most suitable working sensor is selected, thereby realizing intelligent sensor working state adjustment, avoiding the continuous activation of all sensors, and only activating the most relevant sensors as needed. The energy consumption of the pipeline leakage monitoring system is low.

[0067] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A pipeline leakage monitoring method based on infrasound, characterized in that: The following steps are involved: S1: Infrasonic sensors are arranged at equal intervals on the pipeline, and infrasonic signals inside the pipeline are collected by the infrasonic sensors. The infrasonic signals are preprocessed to extract characteristic parameters of the preprocessed infrasonic signals, where the characteristic parameters include frequency characteristics and energy distribution. The characteristic parameters are input into a pre-trained scoring estimation model, and the location of the leakage point is determined based on a time difference positioning algorithm, and the severity score of each leakage point on the pipeline is output; S2: Mark the part between two adjacent preset connection nodes in the pipeline as a sub-pipeline, obtain the historical alarm record of the sub-pipeline, the historical alarm record stores leakage alarm information within a preset period of time, and each leakage alarm information includes a leakage point location and a corresponding severity score, calculate a screening value based on the historical alarm record, and determine whether the sub-pipeline is the first pipeline based on the screening value, wherein the process of determining the first pipeline includes: calculating the screening value , represents the filter value used to determine the first pipeline, Indicates the sub-pipeline The severity rating of the leak when it occurs, Indicates the total number of leaks on the sub-pipeline. is the preset correction factor, for the length of the sub-pipeline, marking the sub-pipeline whose screening value is less than a preset screening value threshold as the first pipeline; S3: sorting the severity scores of the leakage points when leakage occurs on the first pipeline in descending order to obtain a first order, obtaining the first leakage point in the first order, obtaining the distance parameters between the infrasonic wave sensors and the first leakage point, sorting the distance parameters in ascending order to obtain a second order, determining the corresponding infrasonic wave sensors based on the first three distance parameters in the second order, marking them as target sensors, and determining the target line based on the connecting line of the positions of the three target sensors; The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where B represents the omission ratio, M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold. S4: obtaining the target line A and the number of omission ratio calculations a at this time, determining the working sensor based on the target line A and the number of omission ratio calculations a, and controlling the infrasonic wave sensors on the first pipeline other than the working sensor to enter a dormant state, wherein the process of determining the working sensor comprises: connecting the leakage point positions corresponding to the number of omission ratio calculations by straight lines to obtain a leakage line segment, obtaining the starting point and the end point of the leakage line segment, the severity score at the first position closest to the starting point is less than the severity score at the second position closest to the end point; establishing a number axis, the origin of the number axis is the starting point, and the positive direction of the number axis points from the starting point to the end point, and the unit distance of the number axis is a preset value, generating a coordinate point ( , ), Indicates that the distance from the origin on the number axis is The severity score when a leak occurs at the location of the =0 ; Calculate theoretical position , Indicates the total number of leakages occurring on the portion of the target pipeline corresponding to the target line A, determines the infrasonic sensor closest to the theoretical position, uses it as the key sensor, connects the position of the key sensor and the first position to obtain a first sub-segment, and connects the position of the key sensor and the second position to obtain a second sub-segment; respectively uses the first sub-segment and the second sub-segment as leakage segments, and determines the corresponding working sensors.

2. The pipeline leakage monitoring method based on infrasound according to claim 1 is characterized in that: In step S1, the pre-training process of the score estimation model includes: Establishing a database, wherein the database stores characteristic parameters with annotated severity scores; A rating estimation model is established based on the machine learning model, and the rating estimation model is trained and verified based on the database, wherein the rating estimation model is trained using a back propagation algorithm, and the rating estimation model is verified using a five-fold cross validation.

3. The pipeline leakage monitoring method based on infrasound according to claim 2 is characterized in that: The characteristic parameters stored in the database are based on manually annotated severity scores.

4. The pipeline leakage monitoring method based on infrasound according to claim 1 is characterized in that: In step S1, the process of preprocessing the infrasound wave signal includes denoising, filtering and enhancement.

5. The pipeline leakage monitoring method based on infrasound according to claim 1 is characterized in that: In step S4, after the infrasonic wave sensors on the first pipeline other than the working sensor enter the dormant state, the method further includes the following steps: When the duration of any acoustic wave sensor entering the dormant state reaches a preset duration threshold, a new working sensor is re-determined.

6. A pipeline leakage monitoring system based on infrasound, characterized in that: include: An initial module is used to set infrasonic sensors at equal intervals on the pipeline system, collect infrasonic signals inside the pipeline based on the infrasonic sensors, preprocess the infrasonic signals, extract characteristic parameters of the preprocessed infrasonic signals, the characteristic parameters include frequency characteristics and energy distribution, input the characteristic parameters into a pre-trained scoring estimation model, determine the location of the leakage point based on the time difference positioning algorithm, output the severity score of each leakage point on the pipeline, and determine the location of the leakage point based on the time difference positioning algorithm; A screening module is used to mark a pipeline portion between two adjacent preset connection nodes in a pipeline system as a sub-pipeline, obtain a historical alarm record of the sub-pipeline, wherein the historical alarm record stores leakage alarm information within a preset period, wherein a single leakage alarm information includes a leakage point location and a corresponding severity score, calculate a screening value based on the historical alarm record, and determine whether the sub-pipeline is a first pipeline based on the screening value, wherein the process of determining the first pipeline includes: calculating the screening value , represents the filter value used to determine the first pipeline, Indicates the sub-pipeline The severity rating of the leak when it occurs, Indicates the total number of leaks on the sub-pipeline. is the preset correction factor, for the length of the sub-pipeline, marking the sub-pipeline whose screening value is less than a preset screening value threshold as the first pipeline; an optimization module, configured to sort the severity scores of each leakage point when leakage occurs on the first pipeline in descending order to obtain a first order, obtain the first leakage point at the first position in the first order, obtain the distance parameters between each of the infrasonic wave sensors and the first leakage point, sort each of the distance parameters in ascending order to obtain a second order, determine the corresponding infrasonic wave sensor based on the first three distance parameters in the second order, mark it as a target sensor, and determine a target line based on a line connecting the positions of the three target sensors; The portion of the first pipeline corresponding to the target line is taken as the target pipeline, the number of leakage points on the first pipeline other than the target pipeline is obtained, and the omission ratio B=M / m is calculated, where M represents the number of leakage points on the first pipeline other than the target pipeline, and m represents the total number of leakage points on the target pipeline. If the omission ratio is greater than or equal to a preset ratio threshold, the first severity score is removed from the first sorting, a new first sorting is obtained, and a new target sensor is determined. A new target line is determined according to all the target sensors and a new omission ratio is calculated until a new omission ratio is less than the ratio threshold. The adjustment module is used to obtain the target line A and the calculation times a of the missing ratio at this time, determine the working sensor based on the target line A and the calculation times a of the missing ratio, and control the infrasonic wave sensors on the first pipeline other than the working sensor to enter a dormant state, wherein the process of determining the working sensor includes: connecting the leakage point positions corresponding to the calculation times of the missing ratio by straight lines to obtain a leakage line segment, obtaining the starting point and the end point of the leakage line segment, and the severity score at the first position closest to the starting point is less than the severity score at the second position closest to the end point; establishing a number axis, the origin of the number axis is the starting point, and the positive direction of the number axis points from the starting point to the end point, and the unit distance of the number axis is a preset value, generating a coordinate point ( , ), Indicates that the distance from the origin on the number axis is The severity score when a leak occurs at the location of the =0; calculate theoretical position , Indicates the total number of leakages occurring on the portion of the target pipeline corresponding to the target line A, determines the infrasonic sensor closest to the theoretical position, uses it as the key sensor, connects the position of the key sensor and the first position to obtain a first sub-segment, and connects the position of the key sensor and the second position to obtain a second sub-segment; respectively uses the first sub-segment and the second sub-segment as leakage segments, and determines the corresponding working sensors.

Images