CN116906837B

CN116906837B - State monitoring system and monitoring method for underground pipeline

Info

Publication number: CN116906837B
Application number: CN202311189337.7A
Authority: CN
Inventors: 刘华; 张沛良; 曲慧磊; 么远
Original assignee: Shanghai Tongji Engineering Consulting Co ltd
Current assignee: Shanghai Tongji Engineering Consulting Co ltd
Priority date: 2023-09-15
Filing date: 2023-09-15
Publication date: 2023-11-28
Anticipated expiration: 2043-09-15
Also published as: CN116906837A

Abstract

The invention provides a state monitoring system and a monitoring method for an underground pipeline, wherein the system comprises: monitoring device and calculation unit. Wherein the monitoring device comprises an inclination monitoring component, a precipitation monitoring component and a stress monitoring component. The inclination monitoring component is arranged on the periphery of the pipeline and acquires inclination degree information of the pipeline at preset first time intervals. The precipitation monitoring component is arranged on the ground surface and acquires precipitation information of the buried pipe area at a preset second time interval. The stress monitoring component is arranged above the pipeline in the vertical direction, and stress information of soil around the pipeline on the pipeline is acquired at a preset third time interval. The calculation unit calculates pipeline state information of the underground pipeline in a future period of time according to the inclination degree information, the precipitation amount information and the stress information from the inclination monitoring component, the precipitation amount monitoring component and the stress monitoring component. The scheme can accurately judge the state of the underground pipeline according to the acquired information.

Description

State monitoring system and monitoring method for underground pipeline

Technical Field

The invention relates to the technical field of underground pipe network monitoring, in particular to a state monitoring system and a state monitoring method for an underground pipeline.

Background

Urban underground pipelines are an important component of urban infrastructure. With the rapid development of urban construction, the development and utilization of underground space are increased in recent years, and more underground pipelines are available. However, because of the complex buried and archival materials of the underground pipelines, the state of many underground pipelines is not clear, and the hidden property of the underground pipelines is added, the problem that the underground pipelines are damaged in the construction process frequently occurs, which can cause great difficulty to urban construction and old area transformation. Therefore, the state of the underground pipeline directly affects the efficiency and difficulty of urban construction.

Currently, metal pipe detectors are mostly used for detecting the state of underground pipelines. However, because the metal probe is detected by utilizing the principle of electromagnetic signals, the metal probe is easily interfered by stray current and geological environment in the working process, and the detection efficiency is unstable. Moreover, the metal probe can only determine the position, the trend and the sheath fault point of the underground pipeline, and the state of the pipeline cannot be accurately determined only by the parameters.

Therefore, the state of the pipeline cannot be accurately determined by the pipeline monitoring method in the prior art, and the monitoring effect is poor.

Disclosure of Invention

The invention aims to solve the problems that the pipeline state cannot be accurately judged and the pipeline monitoring effect is poor in the prior art.

To solve the above problems, an embodiment of the present invention discloses a condition monitoring system for an underground pipeline, including: monitoring device, monitoring device includes: the inclination monitoring component is arranged on the periphery of the pipeline and used for acquiring inclination degree information of the pipeline at a preset first time interval; the precipitation monitoring component is arranged on the ground surface and acquires precipitation information of the buried pipe area at a preset second time interval; the stress monitoring component is arranged above the pipeline in the vertical direction, and stress information of soil around the pipeline on the pipeline is obtained at a preset third time interval; and the calculating unit is respectively in communication connection with the inclination monitoring component, the precipitation monitoring component and the stress monitoring component, and calculates pipeline state information of the underground pipeline in a future period of time according to inclination degree information, precipitation amount information and stress information from the inclination monitoring component, the precipitation monitoring component and the stress monitoring component.

By adopting the scheme, the inclination monitoring component, the precipitation monitoring component and the stress monitoring component are arranged, so that the inclination of the underground pipeline under the action of soil, the state of the soil, the stress of the soil on the underground pipeline and the like can be accurately reflected, the state of the underground pipeline can be accurately reflected according to the data, and the accuracy of pipeline state measurement is improved. And the pipeline state in a future period is predicted according to the acquired historical data, so that enough maintenance and inspection time can be reserved for maintenance personnel, and the safety of the underground pipeline is further ensured.

According to another embodiment of the present invention, the condition monitoring system for an underground pipeline disclosed in the embodiment of the present invention, the monitoring device further includes: the vibration monitoring component is arranged above the pipeline in the vertical direction and at a position which is a preset distance away from the pipeline, and the vibration monitoring component monitors vibration frequency information of the position where the vibration monitoring component is positioned in real time; the calculation unit is in communication connection with the vibration monitoring component, acquires vibration frequency information from the vibration monitoring component, and generates alarm information according to the vibration frequency information and a preset vibration frequency threshold value.

By adopting the scheme, the vibration monitoring component is arranged at the position above the pipeline by a preset distance, the vibration frequency information of the position where the vibration monitoring component is located is monitored by the vibration monitoring component, and when the vibration frequency information exceeds the vibration frequency threshold value, alarm information is generated to remind maintenance personnel of paying attention to the state of the underground pipeline, and preventive measures are timely taken when other construction actions possibly damage the underground pipeline.

According to another embodiment of the present invention, the condition monitoring system for an underground pipeline disclosed in the embodiment of the present invention, the monitoring device further includes: the image acquisition component is arranged on the ground surface corresponding to the pipeline, and deformation information of the ground surface is acquired at a preset fourth time interval; the computing unit is in communication connection with the image acquisition component and generates alarm information according to pipeline state information and deformation information. The state monitoring system also comprises an alarm unit, wherein the alarm unit is in communication connection with the computing unit, acquires alarm information from the computing unit and alarms according to the alarm information; the alarm information comprises primary alarm information, secondary alarm information and tertiary alarm information; the alarm level of the second-level alarm information is higher than that of the third-level alarm information and lower than that of the first-level alarm information. And the alarm information generated according to the vibration frequency information and the preset vibration frequency threshold value is first-level alarm information.

By adopting the scheme, the image acquisition component is arranged on the ground surface, and the alarm information is generated according to the monitoring result of the image acquisition component and the pipeline state information, so that the environment of a buried pipe area is considered, the condition of the ground surface is considered, and the generated alarm information is more accurate. Further, the classified alarm is carried out, so that maintenance personnel can quickly know the condition of the pipeline, and accordingly appropriate reactions can be made according to the alarms of different degrees, and maintenance efficiency of the pipeline is improved.

According to another embodiment of the present invention, the state monitoring system for an underground pipeline according to the embodiment of the present invention, the calculating unit calculates pipeline state information of the underground pipeline according to inclination degree information, precipitation amount information, and stress information, including: the calculation unit predicts a precipitation curve in a future period of time according to precipitation information, divides the future period of time into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to the precipitation curve and a preset precipitation threshold, and calculates pipeline state information in a first preset time range in each high precipitation time period and pipeline state information in a second preset time range in each low precipitation time period according to inclination degree information and stress information respectively. Wherein the first predetermined time range is a time range of a first tenth of the high precipitation time period; the second predetermined time range is the first half of the low precipitation period. And, the calculation unit generates alarm information according to the pipeline state information and the deformation information, including: the calculating unit determines a state influence coefficient according to the deformation information, calculates an alarm value according to the state influence coefficient and pipeline state information, and determines alarm information according to the alarm value and a preset numerical range. The deformation information comprises the height subsidence of the ground surface; if the surface height subsidence is less than or equal to 0.2 cm/month, the state influence coefficient is 0.5; if the subsidence of the surface height is less than 0.2 cm/month and less than 1 cm/month, the state influence coefficient is 0.8; if the surface height subsidence is more than or equal to 1 cm/month, the state influence coefficient is 1. If the alarm value is in the first numerical range, determining that the alarm information is primary alarm information; if the alarm value is in the second value range, determining that the alarm information is the secondary alarm information; if the alarm value is in the third numerical range, determining that the alarm information is three-level alarm information; if the alarm value is in the fourth numerical range, no alarm information is generated.

By adopting the scheme, the predicted precipitation curve is divided into the high precipitation time period and the low precipitation time period, and when the pipeline state information of the high precipitation time period is calculated, the pipeline state information of the first tenth of the time in the period is calculated, so that the calculation efficiency can be improved, and the alarm efficiency when the pipeline is possibly failed or damaged is further improved. When calculating the pipeline state information of the low-precipitation time period, calculating the pipeline state information of the first half of the time in the low-precipitation time period can improve the calculation accuracy, further improve the accuracy of alarm information generation and prevent false alarm.

According to another embodiment of the present invention, the condition monitoring system for an underground pipeline disclosed in the embodiment of the present invention calculates a predetermined distance between a vibration monitoring part and the pipeline according to the following formula:

wherein H is a preset distance, L is the pipeline burial depth, R is the pipeline radius, E is the pipeline material rigidity, and the parameters are all obtained in international units and are substituted into a formula for calculation; the range of the first time interval is 1-10 days; the second time interval ranges from 1 day to 10 days; the third time interval ranges from 1 day to 10 days; the fourth time interval ranges from 1 day to 10 days; the range of the vibration frequency threshold value is 0.8 Hz-1.2 Hz; the range of the precipitation threshold value is 6 mm-8 mm; the range of the future period of time is 3 months to 6 months; the first numerical range is 1-3; the second numerical range is 4-5; the third numerical range is 6-7; the fourth numerical range is 8-10.

By adopting the scheme, when the vibration monitoring component is arranged, the influences of the pipeline burial depth, the pipeline radius and the pipeline material rigidity are considered, so that the accuracy of the arrangement position of the vibration monitoring component is improved.

The embodiment of the invention discloses a state monitoring method of an underground pipeline, which is applicable to the state monitoring system of the underground pipeline described in any embodiment; and, the state monitoring method includes:

s1: acquiring inclination degree information of a first time period at a first time interval, acquiring stress information of the first time period at a third time interval, and forming a first historical data set; acquiring precipitation information of a second time stage at a second time interval, and forming a second historical data set;

s2: pipeline state information of the underground pipeline in a future period is calculated according to the first historical data set and the second historical data set.

By adopting the scheme, the pipeline state information is calculated according to the information that the inclination degree information, the stress information, the precipitation amount information and the like directly influence the soil state around the pipeline, the influence of the inclination degree, the soil state and the precipitation amount of the pipeline is comprehensively considered, and the pipeline state measurement precision is improved. And the pipeline state in a future period is predicted according to the acquired historical data, so that enough maintenance and inspection time can be reserved for maintenance personnel, and the safety of the underground pipeline is further ensured.

According to another embodiment of the present invention, the method for monitoring the status of an underground pipeline disclosed in the embodiment of the present invention, step S1 further includes:

and obtaining deformation information at a fourth time interval, and determining a state influence coefficient according to the deformation information. And, after step S2, further includes:

s3: and generating alarm information according to the pipeline state information and the state influence coefficient. In addition, in the process of executing the step S1 to the step S3, the state monitoring method further comprises the following steps: monitoring vibration frequency information at a position above the pipeline and at a preset distance from the pipeline in real time, and judging whether the vibration frequency information is larger than a preset vibration frequency threshold value or not;

if yes, generating alarm information;

if not, continuing to judge whether the vibration frequency information is larger than the vibration frequency threshold value.

By adopting the scheme, the alarm information is generated according to the pipeline state information related to the stress, the inclination degree and the like of the pipeline and the state influence coefficient related to the surface height sinking amount, and the environmental information around the pipeline and the environmental information of the surface corresponding to the pipeline are comprehensively considered, so that the accuracy of the alarm information generation can be improved.

According to another embodiment of the present invention, the method for monitoring the condition of an underground pipeline disclosed in the embodiment of the present invention, step S2 includes:

S21: predicting a precipitation curve in a future period of time by using a prediction algorithm according to the second historical data set;

s22: dividing a precipitation curve into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to a preset precipitation threshold; pipeline state information in a first preset time range in each high-precipitation time period and pipeline state information in a second preset time range in each low-precipitation time period are respectively calculated according to the first historical data set by using a regression algorithm. In step S3, the alarm information includes primary alarm information, secondary alarm information and tertiary alarm information; the step S3 comprises the following steps:

s31: performing product operation on the pipeline state information and the state influence coefficient to determine an alarm value;

s32: and determining alarm information according to the alarm value and a preset numerical range. If the alarm value is in the first numerical range, determining that the alarm information is primary alarm information; if the alarm value is in the second value range, determining that the alarm information is the secondary alarm information; if the alarm value is in the third numerical range, determining that the alarm information is three-level alarm information; if the alarm value is in the fourth numerical range, no alarm information is generated. And the alarm information generated when the vibration frequency information is larger than the vibration frequency threshold value is first-level alarm information.

By adopting the scheme, the hierarchical alarm is carried out, and maintenance personnel can carry out corresponding actions according to different information, so that the pipeline maintenance efficiency is improved.

According to another embodiment of the present invention, in the method for monitoring the state of an underground pipeline disclosed in the embodiment of the present invention, in step S21, the prediction algorithm is a long-short-term memory network algorithm; and, step S21 includes:

s211: normalizing the data in the second historical data set to obtain a second data set, and inputting the second data set to an input layer of a long-term and short-term memory network algorithm;

s212: training and modeling the long-term memory network according to the second data set to obtain a precipitation prediction model;

s213: and training a precipitation prediction model, and outputting a precipitation curve by using the trained precipitation prediction model. And, step S22 includes:

s221: taking a section of the precipitation curve, the precipitation value of which is higher than the precipitation threshold value, as a high precipitation time period, and taking a section of the precipitation curve, the precipitation value of which is lower than the precipitation threshold value, as a low precipitation time period;

s222: performing outlier rejection processing on the first historical data set to obtain a first data set;

s223: integrating the relation between the first data and the pipeline state information based on the sampling time into a data group, and randomly dividing the data group into a training set and a testing set based on a preset proportion after the sequence of each data group is disordered;

S224: generating a parameter combination of a k-nearest neighbor regression algorithm, analyzing fitting degrees of the k-nearest neighbor regression algorithm to a training set under different parameter combinations, determining optimal parameters in the parameter combination, and constructing a pipeline state prediction model according to the optimal parameters and the k-nearest neighbor regression algorithm;

s225: the first data is input into a pipeline state prediction model to calculate pipeline state information within a first predetermined time range in each high precipitation time period and pipeline state information within a second predetermined time range in each low precipitation time period.

By adopting the scheme, the regression algorithm is utilized to calculate the pipeline state information, so that the efficiency of calculating the pipeline state information can be improved, and the alarm information can be timely and quickly generated when the pipeline is possibly damaged. And the LSTM prediction model is utilized to predict the precipitation, so that the accuracy of precipitation prediction can be improved.

According to another specific embodiment of the invention, the condition monitoring method of the underground pipeline disclosed by the embodiment of the invention has the range of 3 months to 12 months in the first time period; the range of the second time period is 6 months to 12 months; the alarm level of the second-level alarm information is higher than that of the third-level alarm information and lower than that of the first-level alarm information; the first, second, third, and fourth numerical ranges increase in sequence.

The beneficial effects of the invention are as follows:

according to the underground pipeline state monitoring system, the inclination monitoring component, the precipitation monitoring component and the stress monitoring component are arranged, so that the inclination amount of the underground pipeline under the action of soil, the state of the soil, the stress of the soil on the underground pipeline and the like can be accurately reflected, the state of the underground pipeline can be accurately reflected according to the data, and the pipeline state measuring precision is improved.

Drawings

FIG. 1 is a schematic diagram of a condition monitoring system for an underground pipeline according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a precipitation curve provided by an embodiment of the present invention;

FIG. 3 is a flow chart of a method for monitoring the condition of an underground pipeline according to an embodiment of the present invention;

FIG. 4 is a schematic flow chart of another embodiment of a method for monitoring the condition of an underground pipeline;

FIG. 5 is another flow chart of a method for monitoring the condition of an underground pipeline according to an embodiment of the present invention;

fig. 6 is another flow chart of a method for monitoring the condition of an underground pipeline according to an embodiment of the invention.

Reference numerals illustrate:

1. a monitoring device; 11. a tilt monitoring component; 12. a precipitation amount monitoring unit; 13. a stress monitoring component; 14. a vibration monitoring component; 15. an image acquisition section; 2. a calculation unit; 3. and an alarm unit.

Detailed Description

Example 1:

in order to solve the problems that in the prior art, the pipeline state cannot be accurately judged and the pipeline monitoring effect is poor, an embodiment of the invention provides a state monitoring system for an underground pipeline, and reference is made to fig. 1. The condition monitoring system of the underground pipeline comprises a monitoring device 1 and a computing unit 2. Wherein the monitoring device 1 is used for acquiring information related to the state of the underground pipeline, and the calculating unit 2 is used for calculating the state information of the underground pipeline in a future period according to the data information acquired by the monitoring device 1. Specifically, the monitoring device 1 includes an inclination monitoring part 11, a precipitation monitoring part 12, and a stress monitoring part 13. Wherein the inclination monitoring part 11 is arranged at the periphery of the pipeline, and acquires inclination degree information of the pipeline at a preset first time interval; the precipitation monitoring unit 12 is provided on the ground surface, and acquires precipitation information of the buried pipe region at predetermined second time intervals; the stress monitoring section 13 is disposed above the pipeline in the vertical direction, and acquires stress information of soil around the pipeline on the pipeline at predetermined third time intervals. The calculation unit 2 is in communication connection with the inclination monitoring means 11, the precipitation monitoring means 12, and the stress monitoring means 13, respectively, and calculates pipeline state information of the underground pipeline in a future period of time based on inclination degree information, precipitation amount information, and stress information from the inclination monitoring means 11, the precipitation monitoring means 12, and the stress monitoring means 13.

Specifically, the inclination monitoring part 11 is an inclination sensor for measuring an inclination angle of the pipeline with respect to the horizontal axis. The precipitation monitoring unit 12 is a skip type rain gauge, which is disposed on the earth surface corresponding to the buried pipe region, and is used for measuring the precipitation of the buried pipe region, and typically, one precipitation monitoring unit 12 is disposed every 10 km. The stress monitoring component 13 is a fiber grating stress sensor and is used for measuring the stress of the soil on the pipeline. The inclination monitoring part 11 and the stress monitoring part 13 are both arranged at a position close to the pipeline, and can be integrated into a whole structure. The calculating unit 2 is a single chip microcomputer with data receiving, transmitting and calculating functions, and is connected with the monitoring device 1, and calculates according to the monitoring result of the monitoring device 1 so as to predict pipeline state information representing the state of the underground pipeline in a future period of time.

More specifically, the first time interval ranges from 1 day to 10 days, such as 1 day, 3 days, 6.5 days, 10 days, or other times within the range; the second time interval ranges from 1 day to 10 days, such as 1 day, 3 days, 6.5 days, 10 days, or other times within the range; the third time interval may range from 1 day to 10 days, such as 1 day, 3 days, 6.5 days, 10 days, or other times within the range. Because the surrounding soil or the environment does not rapidly mutate after the underground pipeline is buried underground, the ranges of the first time interval, the second time interval and the third time interval are all set to be 1-10 days, the data do not need to be acquired in real time, the data amount required to be acquired by the monitoring device 1 is reduced, and the data acquisition cost is saved; in addition, the change of the surrounding environment of the underground pipeline can be accurately monitored within the time range of 1-10 days, and an accurate data basis is provided for the calculation of the subsequent calculation unit 2. The range of future time is 3 months to 6 months, such as 3 months, 4 half months, 6 months or other time ranges from the current time. Setting the future period of time to 3 months-6 months, that is, the computing unit 2 can predict the pipeline state of 3 months-6 months in the future, and can provide enough reaction time for maintenance personnel when the pipeline may fail. In addition, the time of 3 months to 6 months is not very long, and the problem of low prediction accuracy caused by overlong time is avoided.

In this way, since the state around the underground pipe is mainly affected by the state of the surrounding soil, the inclination monitoring means 11, the precipitation monitoring means 12, and the stress monitoring means 13 are provided, and the inclination amount of the underground pipe due to the soil, the state of the soil, the stress of the soil on the underground pipe, and the like can be accurately reflected, and the state of the underground pipe can be accurately reflected based on these data, thereby improving the accuracy of the pipe state measurement. And, according to the historical data that obtains, predict the pipeline state of a period of time in the future, but not calculate the pipeline state under the present, when the pipeline state of calculation is predicted to be probably out of order, also can leave the sufficient maintenance inspection time of maintenance personnel. Compared with the pipeline which is processed after the problem occurs, the scheme can prevent the problem, so that the influence of the possible faults of the pipeline on the normal life of people is reduced as much as possible.

Further, in the condition monitoring system of the underground piping according to the present invention, referring to fig. 1, the monitoring apparatus 1 further includes: and a vibration monitoring part 14, the vibration monitoring part 14 being disposed above the pipeline at a predetermined distance from the pipeline in the vertical direction. The vibration monitoring section 14 monitors vibration frequency information at the position where it is located in real time. The calculation unit 2 is connected to the vibration monitoring unit 14 in a communication manner, acquires vibration frequency information from the vibration monitoring unit 14, and generates alarm information based on the vibration frequency information and a preset vibration frequency threshold value.

Specifically, the vibration monitoring component 14 is a resistance strain type vibration sensor that can measure the mechanical vibration frequency at a set position. In this embodiment, when the vibration frequency information monitored by the vibration monitoring component 14 is greater than or equal to the preset vibration frequency threshold value, it is indicated that the soil around the vibration monitoring component 14 is severely vibrated, and the vibration may be caused by other construction actions, so as to avoid damage to the underground pipeline arranged below the vibration monitoring component 14 by the construction actions, and generate alarm information. More specifically, the vibration frequency threshold ranges from 0.8Hz to 1.2Hz, such as 0.8Hz, 1Hz, 1.2Hz, or other frequency values within the range.

In this way, the vibration monitoring component 14 is further arranged above the pipeline at a preset distance, vibration frequency information of the position where the vibration monitoring component 14 is located is monitored by utilizing the vibration monitoring component 14, when the vibration frequency information exceeds a vibration frequency threshold value, alarm information is generated to remind maintenance personnel of paying attention to the state of the underground pipeline, and preventive measures are timely taken when other construction actions possibly damage the underground pipeline. In addition, the vibration monitoring component 14 monitors the vibration frequency information in real time, so that alarm information can be generated at the first time when the underground pipeline is possibly damaged, and the subsequent maintenance efficiency is improved. The vibration monitoring component 14 is positioned a distance above the pipeline to generate an alarm message in time when no substantial damage has been made to the pipeline.

Still further, in the condition monitoring system of the underground piping according to the present invention, the predetermined distance of the vibration monitoring section 14 from the piping is calculated according to the following formula:

wherein H is a preset distance, L is the pipeline burial depth, R is the pipeline radius, E is the pipeline material rigidity, and the parameters are all obtained in international units and are substituted into a formula for calculation. That is, in calculating the predetermined distance using the above formula, calculation is performed using only specific values of L, R, E. For example, the pipeline burial depth L, the pipeline radius R and the predetermined distance H are all taken as units of meters, and the pipeline material stiffness E is taken as units of Mpa.

Specifically, the pipeline burial depth L is generally below 1m, the approximate range is 1m to 5m, and the pipeline radius R is generally 0.35m to 1m; the rigidity of the pipeline material is determined according to the specific material adopted by the pipeline material, the rigidity of the material of the pp pipe is approximately 50Mpa, and the rigidity of the metal material is generally 200Mpa. According to the above formula, a metal pipe with a burial depth of 2m and a pipeline radius of 0.5m is buried, and the vibration monitoring part 14 is arranged at a position of 0.23m at the top of the pipeline; the vibration monitoring section 14 was placed at a position of 0.53m on the top of the pipeline with a pp pipe buried 3m deep and having a pipeline radius of 0.7 m. That is, the greater the stiffness of the pipeline, the deeper the burial depth, the closer the vibration monitoring component 14 is disposed to the pipeline, and the greater the radius of the pipeline, the further the vibration monitoring component 14 is disposed to the pipeline. In this way, the accuracy of the position of the vibration monitoring component 14 is improved in consideration of the effects of the pipe burial depth, the pipe radius, and the pipe material stiffness when the vibration monitoring component 14 is provided.

Further, in the condition monitoring system of the underground piping according to the present invention, referring to fig. 1, the monitoring apparatus 1 further includes: an image acquisition section 15. The image acquisition unit 15 is provided on the surface corresponding to the pipeline, and acquires the deformation information of the surface at predetermined fourth time intervals. The calculation unit 2 is connected to the image acquisition unit 15 in communication, and generates alarm information based on the pipeline state information and the strain information. Specifically, the image capturing section 15 is a camera that obtains photographs of the earth's surface at predetermined fourth time intervals, and calculates deformation information of the earth's surface using an image analysis method. And the calculation unit generates alarm information together according to the pipeline state information and the deformation information so as to timely inform maintenance personnel under the condition that the pipeline is possibly damaged. More specifically, the fourth time interval ranges from 1 day to 10 days, such as 1 day, 3 days, 6.5 days, 10 days, or other times within the range. The fourth time interval is set to be 1-10 days, so that a good balance between data acquisition efficiency and accuracy can be achieved. In this way, by providing the image acquisition unit 15 on the earth's surface and generating the alarm information based on the monitoring result of the image acquisition unit 15 and the pipeline status information, not only the environment of the buried pipe area but also the situation of the earth's surface are considered, so that the generated alarm information is more accurate.

Further, in the condition monitoring system for an underground pipeline according to the present invention, referring to fig. 1, the condition monitoring system further includes an alarm unit 3, the alarm unit 3 is communicatively connected with the calculation unit 2, acquires alarm information from the calculation unit 2, and alarms according to the alarm information. The alarm information comprises primary alarm information, secondary alarm information and tertiary alarm information; the alarm level of the second-level alarm information is higher than that of the third-level alarm information and lower than that of the first-level alarm information. In particular, the alarm unit 3 includes, but is not limited to, an alarm bell or indicator light provided at the location of the maintenance personnel. For example, the three-level alarm information can be that the alarm rings for 1s, and the indicator light is green, and the interval 10s is lighted once; the secondary alarm can be that the alarm rings for 3s, and the indicator light is a yellow light which is lighted once at intervals of 3 s; the primary alarm can be that the alarm rings for 10s, and the indicator light is red light and continuously lights up. The method has the advantages that the grading alarm is carried out according to the alarm information, so that maintenance personnel can quickly know the severity of the possible damage of the pipeline, and the maintenance personnel can conveniently take proper maintenance measures.

In this embodiment, the alarm information generated according to the vibration frequency information and the preset vibration frequency threshold is first-level alarm information. That is, if the vibration frequency of the vibration monitoring unit 14 at the installation position is high, it is indicated that some construction operations may be performed, and damage may be caused to the buried underground pipeline in a short time due to the large action of the construction operations, at this time, primary alarm information may be directly generated to remind maintenance personnel to take corresponding measures promptly.

Further, in the condition monitoring system of an underground pipeline according to the present invention, the calculating unit 2 calculates pipeline condition information of the underground pipeline based on the inclination degree information, the precipitation amount information, and the stress information, including: the calculating unit 2 predicts a precipitation curve in a future period of time according to precipitation information, and divides the future period of time into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to the precipitation curve and a preset precipitation threshold. The calculation unit calculates pipeline state information in a first predetermined time range in each high-precipitation time period and pipeline state information in a second predetermined time range in each low-precipitation time period according to the inclination degree information and the stress information. Specifically, the first predetermined time range is a time range of the first tenth of the high precipitation time period; the second predetermined time range is the first half of the low precipitation period. The precipitation threshold value ranges from 6mm to 8mm, for example 6mm, 6.5mm, 7mm, 8mm, or other values within this range.

More specifically, an example will be described in which the future period of time is 3 months and the precipitation threshold is 8 mm. Assume that the precipitation situation for the next 3 months is as shown in fig. 2, wherein the period exceeding 8mm is a high precipitation period and the period not exceeding 8mm is a low precipitation period. The precipitation curve shown in the figure includes 4 times of high precipitation: day 13 to 16, 24 to 27, 36 to 41, 64 to 75; low precipitation time period 5: day 1 to 12, 17 to 23, 28 to 35, 42 to 63, 76 to 90. Then in calculating the pipeline state information, it is necessary to calculate the pipeline state information for the first tenth of the time period of high precipitation and the pipeline state information for the first half of the time period of low precipitation to represent the pipeline state information for a future period of time using the pipeline state information for the first tenth of the time period of high precipitation and the pipeline state information for the first half of the time period of low precipitation. For example, pipeline state information in the time range of 0 to 36 minutes at 9 am on day 13, and pipeline state information on days 1 to 6 are calculated. In this embodiment, the time for alarming according to the alarm information may be 1 to 10 days before the corresponding time of the high precipitation time period and the low precipitation time period. For example, after calculating the pipeline status information from day 13 to day 16 from today, an alarm may be given on day 10 from today (i.e., the first 3 days of the high precipitation period). The time of alarming is not too early or too late, so that the situation that maintenance personnel forget due to early alarming or the maintenance is not completed due to too late alarming is avoided.

In this way, the predicted precipitation curve is divided into the high precipitation time period and the low precipitation time period, and the probability that soil will change is high because of the large precipitation in the high precipitation time period, so that when calculating pipeline state information, only pipeline state information of the first tenth of the time in the period is calculated, the calculation efficiency can be improved, and the alarm efficiency when the pipeline is possibly faulty or damaged is further improved. In the low precipitation time period, the precipitation amount is smaller, and the possibility that soil can change is smaller, so that when pipeline state information is calculated, the pipeline state information of the first half time in the period is calculated, the calculation accuracy can be improved, the generation accuracy of alarm information is further improved, and false alarm is prevented.

Further, in the condition monitoring system of an underground pipeline according to the present invention, the calculating unit generates alarm information based on the pipeline condition information and the deformation information, including: the calculating unit determines a state influence coefficient according to the deformation information, calculates an alarm value according to the state influence coefficient and pipeline state information, and determines alarm information according to the alarm value and a preset numerical range. Specifically, the deformation information includes the amount of surface height subsidence. If the surface height subsidence is less than or equal to 0.2 cm/month, the state influence coefficient is 0.5; if the subsidence of the surface height is less than 0.2 cm/month and less than 1 cm/month, the state influence coefficient is 0.8; if the surface height subsidence is more than or equal to 1 cm/month, the state influence coefficient is 1. More specifically, the alarm value is obtained by multiplying the state influence coefficient by the pipeline state information.

Further, in the condition monitoring system for an underground pipeline according to the present invention, if the alarm value is in the first numerical range, the alarm information is determined to be the primary alarm information; if the alarm value is in the second value range, determining that the alarm information is the secondary alarm information; if the alarm value is in the third numerical range, determining that the alarm information is three-level alarm information; if the alarm value is in the fourth numerical range, no alarm information is generated. Specifically, the first numerical range is 1-3; the second numerical range is 4-5; the third numerical range is 6-7; the fourth numerical range is 8-10.

More specifically, in this embodiment, the pipeline state information calculated by the calculating unit according to the inclination degree information, the precipitation amount information, and the stress information has a value of 0 to 10. Assuming that the height subsidence amount of the ground surface is 0.18 cm/month, the pipeline state information is 8, the state influence coefficient is 0.5 at the moment, the alarm value determined according to the pipeline state and the state influence coefficient is 4, and the second-level alarm is carried out at the moment in a second value range.

Example 2:

based on the above-mentioned system for monitoring the condition of an underground pipeline, the present embodiment also provides a method for monitoring the condition of an underground pipeline, which is applicable to the system for monitoring the condition of an underground pipeline described in the above-mentioned embodiments. Specifically, referring to fig. 3, the status monitoring method includes:

Specifically, the inclination monitoring means is used to acquire inclination information of the pipeline at first time intervals, for example, inclination information acquired every two days for a total of 3 months. Stress information of the first time period is acquired at third time intervals by using the stress monitoring component, for example, stress information is acquired every two days, and the total stress information is acquired for 3 months. The acquired inclination degree information and stress information of 3 months are formed into a first historical data set. Then, precipitation information of a second time period is acquired at a second time interval by the precipitation monitoring unit, for example, precipitation information is acquired once a day, precipitation information for 6 months is acquired in total, and the acquired precipitation information for 6 months is formed into a second history data set (step S1). Next, pipeline state information of the underground pipeline for a period of time in the future is predicted from the history data of the first history data set and the second history data set (step S2).

More specifically, the first time period ranges from 3 months to 12 months; for example, it may be 3 months, 5 months, 10 months, 12 months, or other times within this range; the range of the second time period is 6 months to 12 months; for example, it may be 6 months, 8 half months, 12 months, or other times within this range. The range of the first time period is set to 3 months to 12 months, and the probability of the soil to be changed drastically in a certain period of time is relatively small, so that the time of 3 months to 12 months is enough to reflect the state change of the soil, the data amount is not too large, the calculation efficiency is not influenced, and the calculation accuracy is not too small. And, setting the range of the second time period relatively long can acquire sufficient precipitation information, so that the accuracy of the subsequent precipitation curve prediction can be improved. And pipeline state information is calculated according to the information that the inclination degree information, the stress information, the precipitation amount information and the like directly influence the soil state around the pipeline, the influence of the inclination degree, the soil state and the precipitation amount of the pipeline is comprehensively considered, and the pipeline state measurement accuracy is improved. And the pipeline state in a future period is predicted according to the acquired historical data, so that enough maintenance and inspection time can be reserved for maintenance personnel, and the safety of the underground pipeline is further ensured.

Further, in the condition monitoring method of the underground pipeline according to the present invention, referring to fig. 4, step S1 further includes:

and obtaining deformation information at a fourth time interval, and determining a state influence coefficient according to the deformation information.

Specifically, the deformation information is acquired at fourth time intervals by the image acquisition means. The fourth time interval ranges from 1 day to 10 days. The deformation information is the surface height subsidence amount. If the height sinking amount of the ground surface is less than or equal to 0.2 cm/month, the state influence coefficient is 0.5; if the subsidence of the surface height is less than 0.2 cm/month and less than 1 cm/month, the state influence coefficient is 0.8; if the surface height subsidence is more than or equal to 1 cm/month, the state influence coefficient is 1.

Further, in the condition monitoring method of an underground pipeline according to the present invention, referring to fig. 4, after step S2, further includes: s3: and generating alarm information according to the pipeline state information and the state influence coefficient.

Specifically, the alarm information is generated based on the pipeline state information related to the stress, inclination degree and the like received by the pipeline and the state influence coefficient related to the surface height sinking amount, and the environmental information around the pipeline and the environmental information of the surface corresponding to the pipeline are comprehensively considered, so that the accuracy of the alarm information generation can be improved.

Further, in the method for monitoring the state of the underground pipeline according to the present invention, in the process of executing the step S1 to the step S3, the method for monitoring the state further includes: monitoring vibration frequency information at a position above the pipeline and at a preset distance from the pipeline in real time, and judging whether the vibration frequency information is larger than a preset vibration frequency threshold value or not;

if yes, generating alarm information;

Specifically, the calculation manner of the predetermined distance and the range of the vibration frequency threshold are described in the embodiments, and the embodiments are not repeated. By the method, the vibration frequency information at the preset distance above the pipeline is monitored in real time, and when the vibration frequency information exceeds the vibration frequency threshold value, alarm information can be timely generated to remind maintenance personnel of paying attention to the state of the underground pipeline, so that damage to the pipeline due to construction actions is avoided.

Further, in the condition monitoring method of the underground piping according to the present invention, referring to fig. 5, step S2 includes:

S22: dividing a precipitation curve into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to a preset precipitation threshold; pipeline state information in a first preset time range in each high-precipitation time period and pipeline state information in a second preset time range in each low-precipitation time period are respectively calculated according to the first historical data set by using a regression algorithm.

Further, in the method for monitoring the condition of the underground pipeline according to the present invention, in step S21, the prediction algorithm is a long-term memory network algorithm. Specifically, step S21 includes:

s213: and training a precipitation prediction model, and outputting a precipitation curve by using the trained precipitation prediction model.

Specifically, in step S211:

is +.>Dimension sample data->Normalized values, ++>And->Respectively->From this, the result after normalization processing of the input time series data can be obtained:

Assume thatIndicating that the hidden layer LSTM cell is +.>Input of time of day->，/>，，…，/>；/>，/>，/>And->Representing a cyclic layer weight matrix,/->，/>，/>Andrepresenting an input layer weight matrix,/->，/>，/>And->Representing the bias vector; />Representing the output vector of the LSTM cell, i.e. the hidden layer state.

In step S213, training the LSTM prediction model by using the data of the current sampling time of the preset length period; calculating a sub-error according to the performance index, judging whether the training error is larger than the sub-error, if so, performing rolling optimization on the LSTM prediction model according to the sub-error, and continuously training the optimized LSTM prediction model with the data of the sampling time of the next preset length segment; otherwise, training the data of the whole sampling time by using the current LSTM prediction model, comparing the training error with the maximum allowable error, if the training error is within the maximum allowable error range, completing the training, otherwise, returning to the step of training the LSTM prediction model by using the data of the sampling time of the current preset length section.

Specifically, the calculation process of LSTM is:

the input gate filters new information of the input signal:

in the method, in the process of the invention,is->The function is activated.

Calculating an alternate cell state：

In the method, in the process of the invention,as a hyperbolic tangent function.

The forget gate discards non-critical information in the history information:

updating cell state, i.e. old cell stateUpdate to New cell State->：

In the method, in the process of the invention,the representation matrix is multiplied by elements.

An output section outputting a state of the gate decision unit:

the current cell state is determined by the tanh functionPartial selection is carried out, and output information is determined by combining an output gate:

the full connection layer outputs predicted values:

in the method, in the process of the invention,for outputting weight matrix +.>Is->In the hidden state of the time, the time is longer,namely LSTM neural network pairThe predicted value of the precipitation amount at the moment is inversely normalized to obtain a final predicted result:

wherein,

in the method, in the process of the invention,to predict the amount of precipitation +.>Dimension sample data->And carrying out inverse normalization on the values. />

Still further, in the condition monitoring method of an underground pipeline according to the present invention, step S22 includes:

The method comprises the steps of calculating the pipeline state information by using the regression algorithm, and can improve the efficiency of calculating the pipeline state information due to the characteristic of high calculation speed of the regression algorithm, so that the alarm information can be timely and quickly generated when the pipeline is possibly damaged. And the LSTM prediction model is used for predicting the precipitation, and the LSTM model has the characteristic of high prediction precision, so that the accuracy of precipitation prediction can be improved.

Further, in the condition monitoring method of an underground pipeline according to the present invention, the alarm information includes primary alarm information, secondary alarm information, and tertiary alarm information in step S3. The alarm level of the second-level alarm information is higher than that of the third-level alarm information and lower than that of the first-level alarm information.

Further, in the condition monitoring method of the underground piping according to the present invention, referring to fig. 6, step S3 includes:

s32: and determining alarm information according to the alarm value and a preset numerical range.

If the alarm value is in the first numerical range, determining that the alarm information is primary alarm information; if the alarm value is in the second value range, determining that the alarm information is the secondary alarm information; if the alarm value is in the third numerical range, determining that the alarm information is three-level alarm information; if the alarm value is in the fourth numerical range, no alarm information is generated. Specifically, the first numerical range, the second numerical range, the third numerical range, and the fourth numerical range increase in order.

Further, in the condition monitoring method of an underground pipeline according to the present invention, the alarm information generated when the vibration frequency information is greater than the vibration frequency threshold value is first-level alarm information.

Next, a specific method of monitoring the condition of an underground pipeline is described.

Firstly, arranging an inclination monitoring component on the periphery of a pipeline, collecting inclination degree information of the pipeline every 2 days, and continuously obtaining the inclination degree information for 3 months; setting a precipitation monitoring component on the ground surface, collecting inclination degree information of a pipeline once every 1 day, and continuously obtaining for 6 months; the stress monitoring component is arranged above the pipeline, stress information of soil around the pipeline on the pipeline is collected every 2 days, and the stress information is continuously obtained for 3 months; setting a vibration monitoring part at a position with a preset distance above a pipeline, and collecting vibration frequency information of the position of the vibration monitoring part in real time; the image acquisition component is arranged on the ground surface of the pipeline embedded region, deformation information of the ground surface is acquired every 2 days, and the deformation information is continuously acquired for 1 month. And integrating the acquired inclination degree information and stress information into a first historical data set, and integrating precipitation amount information into a second historical data set.

Next, a calculation unit in communication with each of the monitoring units acquires data collected by each of the units, and predicts a precipitation curve for 3 months in the future using the LSTM neural network based on the data of the second historical data set.

Then, the calculation unit uses a precipitation threshold of 8mm as a boundary according to the precipitation curve, takes a time period corresponding to the precipitation curve which is equal to or exceeds 8mm as a high precipitation time period, and takes a time period corresponding to the precipitation curve which is less than 8mm as a low precipitation time period.

Next, the calculating unit calculates pipeline state information of the first tenth of the time of the high precipitation time period according to the first historical data set and by using a regression algorithm for each high precipitation time period; for each low precipitation period, pipeline state information for the first half of the low precipitation period is calculated from the first historical data set and using a regression algorithm. Meanwhile, the height subsidence of the ground surface is calculated according to the deformation information acquired by the image acquisition component, and the state influence coefficient is determined according to the height subsidence of the ground surface.

And then, the computing unit multiplies the state influence coefficient and pipeline state information to obtain an alarm value, and generates alarm information of different levels according to a numerical range section where the alarm value is located. It should be noted that, although the alarm value is calculated only based on the pipeline state information of the first tenth of the period of high precipitation or the pipeline state information of the first half of the period of low precipitation, the pipeline state information is calculated based on the first historical data set and using the regression algorithm for the whole period of high precipitation or the whole period of low precipitation.

And, when the vibration frequency acquired by the vibration monitoring part exceeds 1 Hz, the calculation unit directly generates the highest-level alarm information.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing is a further detailed description of the invention with reference to specific embodiments, and it is not intended to limit the practice of the invention to those descriptions. Various changes in form and detail may be made therein by those skilled in the art, including a few simple inferences or alternatives, without departing from the spirit and scope of the present invention.

Claims

1. A condition monitoring system for an underground utility, comprising: a monitoring device, a computing unit;

the monitoring device includes: an inclination monitoring part, a precipitation monitoring part and a stress monitoring part;

the inclination monitoring component is arranged on the periphery of the pipeline and acquires inclination degree information of the pipeline at a preset first time interval;

the precipitation monitoring component is arranged on the ground surface and acquires precipitation information of the buried pipe area at a preset second time interval;

the stress monitoring component is arranged above the pipeline in the vertical direction, and stress information of soil around the pipeline on the pipeline is obtained at a preset third time interval;

The calculation unit is respectively in communication connection with the inclination monitoring component, the precipitation amount monitoring component and the stress monitoring component, predicts a precipitation amount curve in a future period of time according to precipitation amount information from the precipitation amount monitoring component, divides the future period of time into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to the precipitation amount curve and a preset precipitation amount threshold, and respectively calculates pipeline state information in a first preset time range in each high precipitation time period and pipeline state information in a second preset time range in each low precipitation time period according to inclination degree information from the inclination monitoring component and stress information from the stress monitoring component; wherein the method comprises the steps of

The first predetermined time range is a time range of a first tenth of the high precipitation time period;

the second predetermined time range is a time range of a first half of the low precipitation time period; and is also provided with

The monitoring device further includes:

the vibration monitoring component is arranged above the pipeline in the vertical direction and at a position which is a preset distance away from the pipeline, and the vibration monitoring component monitors vibration frequency information of the position where the vibration monitoring component is positioned in real time;

The calculation unit is in communication connection with the vibration monitoring component, acquires vibration frequency information from the vibration monitoring component, and generates alarm information according to the vibration frequency information and a preset vibration frequency threshold value; and is also provided with

The monitoring device further includes:

the image acquisition component is arranged on the ground surface corresponding to the pipeline, and deformation information of the ground surface is acquired at a preset fourth time interval;

the computing unit is in communication connection with the image acquisition component and generates the alarm information according to the pipeline state information and the deformation information; and is also provided with

The state monitoring system further comprises an alarm unit, wherein the alarm unit is in communication connection with the computing unit, acquires the alarm information from the computing unit and alarms according to the alarm information; wherein the method comprises the steps of

The alarm information comprises primary alarm information, secondary alarm information and tertiary alarm information;

the alarm level of the secondary alarm information is higher than that of the tertiary alarm information and lower than that of the primary alarm information; and is also provided with

And the alarm information generated according to the vibration frequency information and a preset vibration frequency threshold value is the primary alarm information.

2. The condition monitoring system of an underground pipeline as recited in claim 1, wherein,

the calculation unit generates the alarm information according to the pipeline state information and the deformation information, and the calculation unit comprises the following steps:

the calculation unit determines a state influence coefficient according to the deformation information, performs product operation on the pipeline state information and the state influence coefficient to determine an alarm value, and determines the alarm information according to the alarm value and a preset numerical range; wherein the method comprises the steps of

The deformation information comprises the height subsidence of the ground surface;

if the surface height subsidence is less than or equal to 0.2 cm/month, the state influence coefficient is 0.5;

if 0.2 cm/month is less than the ground surface height subsidence amount is less than 1 cm/month, the state influence coefficient is 0.8;

if the surface height sinking amount is more than or equal to 1 cm/month, the state influence coefficient is 1; and is also provided with

If the alarm value is in the first numerical range, determining that the alarm information is the primary alarm information;

if the alarm value is in the second value range, determining that the alarm information is the secondary alarm information;

if the alarm value is in the third numerical range, determining that the alarm information is the three-level alarm information;

If the alarm value is in the fourth numerical range, the alarm information is not generated; wherein the method comprises the steps of

The first numerical range is 1-3;

the second numerical range is 4-5;

the third numerical range is 6-7;

the fourth numerical range is 8 to 10.

3. The condition monitoring system of an underground utility line according to claim 2, wherein the predetermined distance of the vibration monitoring component from the utility line is calculated according to the following equation:

wherein H is a preset distance, L is the pipeline burial depth, R is the pipeline radius, E is the pipeline material rigidity, and the parameters are all obtained in international units and are substituted into a formula for calculation; and is also provided with

The first time interval ranges from 1 day to 10 days;

the second time interval ranges from 1 day to 10 days;

the third time interval ranges from 1 day to 10 days;

the fourth time interval ranges from 1 day to 10 days;

the range of the vibration frequency threshold value is 0.8 Hz-1.2 Hz;

the precipitation threshold value ranges from 6mm to 8mm;

the future period of time ranges from 3 months to 6 months.

4. A method for monitoring the condition of an underground pipeline, which is suitable for the condition monitoring system of the underground pipeline according to claim 3; and is also provided with

The state monitoring method comprises the following steps:

s1: acquiring the inclination degree information of a first time period at the first time interval, acquiring the stress information of the first time period at the third time interval, and forming a first historical data set; acquiring the precipitation information of a second time stage at the second time interval, and forming a second historical data set;

s2: the pipeline state information of the underground pipeline in the future period is calculated according to the first historical data set and the second historical data set.

5. The method for monitoring the condition of an underground pipeline according to claim 4, wherein the step S1 further comprises:

acquiring the deformation information at the fourth time interval, and determining the state influence coefficient according to the deformation information; and is also provided with

After the step S2, the method further includes:

s3: generating the alarm information according to the pipeline state information and the state influence coefficient; and is also provided with

In the process of executing the step S1 to the step S3, the state monitoring method further includes:

monitoring vibration frequency information at a position above the pipeline and at a preset distance from the pipeline in real time, and judging whether the vibration frequency information is larger than a preset vibration frequency threshold value or not;

If yes, generating the alarm information;

6. The method for monitoring the condition of an underground pipeline according to claim 5, wherein the step S2 comprises:

s21: predicting a precipitation curve within the future period of time using a prediction algorithm based on the second historical dataset;

s22: dividing the precipitation curve into a plurality of high precipitation time periods and a plurality of low precipitation time periods according to a preset precipitation threshold; calculating the pipeline state information in a first preset time range in each high-precipitation time period and the pipeline state information in a second preset time range in each low-precipitation time period respectively according to the first historical data set by using a regression algorithm; and is also provided with

In the step S3, the alarm information includes primary alarm information, secondary alarm information and tertiary alarm information;

the step S3 includes:

s32: determining the alarm information according to the alarm value and a preset numerical range; wherein the method comprises the steps of

if the alarm value is in the fourth numerical range, the alarm information is not generated; and is also provided with

And the alarm information generated when the vibration frequency information is larger than the vibration frequency threshold value is first-level alarm information.

7. The method for monitoring the condition of an underground pipeline according to claim 6, wherein in the step S21, the prediction algorithm is a long-term and short-term memory network algorithm; and is also provided with

The step S21 includes:

s213: training the precipitation prediction model, and outputting the precipitation curve by using the trained precipitation prediction model; and is also provided with

The step S22 includes:

s221: taking a section of the precipitation curve, the precipitation value of which is higher than the precipitation threshold value, as the high precipitation time period, and taking a section of the precipitation curve, the precipitation value of which is lower than the precipitation threshold value, as the low precipitation time period;

s223: integrating the relation between the first data and the pipeline state information based on sampling time into a data group, and randomly dividing the data group into a training set and a testing set based on a preset proportion after the sequence of each data group is disordered;

s224: generating a parameter combination of a k-nearest neighbor regression algorithm, analyzing the fitting degree of the k-nearest neighbor regression algorithm to the training set under different parameter combinations, determining an optimal parameter in the parameter combination, and constructing a pipeline state prediction model according to the optimal parameter and the k-nearest neighbor regression algorithm;

s225: the first data is input to the pipeline state prediction model to calculate the pipeline state information over a first predetermined time range in each of the high precipitation time periods and the pipeline state information over a second predetermined time range in each of the low precipitation time periods.

8. The method for monitoring the condition of an underground pipeline according to claim 7,

the range of the first time period is 3 months to 12 months;

the range of the second time period is 6 months to 12 months;

the alarm level of the secondary alarm information is higher than that of the tertiary alarm information and lower than that of the primary alarm information;

the first, second, third, and fourth numerical ranges increase in order.