Method and device for identifying major defects of sewage pipe network, electronic equipment and storage medium

Abstract

The application provides a sewage pipe network major defect identification method and device, electronic equipment and storage medium. The method comprises the following steps: acquiring a plurality of first to-be-identified areas of a to-be-identified city, a multivariate data set and a pipe network defect prediction model, inputting the multivariate data set into the pipe network defect prediction model, controlling the pipe network defect prediction model to perform primary defect prediction on the plurality of first to-be-identified areas, and obtaining a first target area; then controlling the pipe network defect prediction model to perform secondary defect prediction on a plurality of second to-be-identified areas in the first target area, obtaining a second target area, performing differential defect identification on the second target area, and obtaining a target defect pipe and a major defect of the target defect pipe. The method of the application effectively improves the positioning accuracy and identification efficiency of the major defect of the pipe network, avoids the limitations of the traditional plugging and water adjustment operation, improves the safety and reliability of the pipe network system, and realizes real-time monitoring of the health status of the sewage pipe throughout its life cycle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewage pipe network detection, and in particular to a sewage pipe network major defect identification method and device, an electronic device and a storage medium. BACKGROUND

[0002] With the acceleration of urbanization, the increase of urban population and the expansion of urban construction, the pressure on urban sewage treatment systems is gradually increasing, and urban sewage pipe systems also face the risk of serious defects such as pipe aging, corrosion, rupture, collapse, structural damage, etc., making the maintenance and management of sewage pipe systems more difficult.

[0003] Serious defects that may occur in sewage pipe systems not only directly affect the sewage discharge function, but also may cause serious public safety incidents such as environmental pollution and ground collapse. In the prior art, the technical means for detecting sewage pipes include CCTV (Closed Circuit Television), sonar detection, pipe periscope detection, ground penetrating radar and other methods. These technical means can be used in combination to detect and evaluate the condition of sewage pipes to ensure the normal operation and maintenance of the pipe system.

[0004] However, the existing sewage pipe detection technology has certain limitations, for example, traditional CCTV (Closed Circuit Television) detection can directly display the internal condition of the pipe, but it is inefficient in large-area investigation, requires water blocking operation during detection, has high detection cost and low efficiency, and cannot adapt to some pipe detection tasks with high water level operation and difficult water blocking. In addition, physical detection technologies such as sonar detection and ground penetrating radar can assist in identifying defects to a certain extent, but the detection sensitivity and positioning accuracy for specific types of major defects are still insufficient. SUMMARY

[0005] The present application provides a sewage pipe network major defect identification method, device, electronic device and storage medium to solve the technical problems of insufficient detection sensitivity and positioning accuracy of existing pipe defect detection technology for major defects, and low pipe defect detection efficiency.

[0006] In a first aspect, the present application provides a sewage pipe network major defect identification method, comprising:

[0007] Obtaining a plurality of first to-be-identified regions of a to-be-identified city, a multivariate data set and a pipe network defect prediction model, the multivariate data set being used to indicate a collection of multi-dimensional geographic information data of the to-be-identified city, and the first to-be-identified region comprising a plurality of second to-be-identified regions;

[0008] controlling the pipe network defect prediction model to perform primary defect prediction on the plurality of first to-be-identified regions according to the multi-element data set, to obtain a first target region, the target region being used to indicate a region in the to-be-identified city where a major pipe network defect exists;

[0009] controlling the pipe network defect prediction model to perform secondary defect prediction on a plurality of second to-be-identified regions in the first target region, to obtain a second target region of the to-be-identified city;

[0010] performing differential defect identification on the second target region, to obtain a target defect pipeline and a major defect of the target defect pipeline.

[0011] Optionally, before the obtaining of the multi-element data set of the to-be-identified city, the first to-be-identified region, and the pipe network defect prediction model, the method further comprises:

[0012] obtaining first geographic information and second geographic information, the geographic information being used to perform multi-level division on the first to-be-identified region;

[0013] performing regional division on the to-be-identified city according to the first geographic information, to obtain the plurality of first to-be-identified regions;

[0014] performing regional division on the plurality of first to-be-identified regions respectively according to the second geographic information, to obtain the plurality of second to-be-identified regions.

[0015] Optionally, the controlling of the pipe network defect prediction model to perform primary defect prediction on the plurality of first to-be-identified regions according to the multi-element data set, to obtain a first target region, comprises:

[0016] obtaining a first region identifier of each of the plurality of first to-be-identified regions, and performing classification processing on the multi-element data set according to the plurality of first region identifiers, to obtain a plurality of first region data sets, the region identifier being used to distinguish data information of the plurality of to-be-identified regions;

[0017] controlling the pipe network defect prediction model to perform defect prediction on the plurality of first region data sets respectively, to obtain a first defect prediction value of each of the plurality of first to-be-identified regions, the defect prediction value being used to indicate a proportion of a major pipe network defect in different regions of the to-be-identified city;

[0018] obtaining a defect prediction threshold of the to-be-identified city, and performing primary screening processing on the plurality of first defect prediction values according to the defect prediction threshold, to obtain the first target region, the defect prediction threshold being used to indicate a lower limit value of the proportion of the major pipe network defect in the to-be-identified region.

[0019] Optionally, the control of the pipe network defect prediction model to the plurality of second to-be-identified regions in the first target region is for secondary defect prediction, and the second target region of the to-be-identified city is obtained.

[0020] Second region identifiers of the plurality of second to-be-identified regions in the first target region are respectively acquired.

[0021] The first region data set of the first target region is classified according to the second region identifiers, and a second region data set of the plurality of second to-be-identified regions in the first target region is obtained.

[0022] The plurality of second region data sets are input into the pipe network defect prediction model, and second defect prediction values of the plurality of second to-be-identified regions in the first target region are obtained.

[0023] The plurality of second defect prediction values are secondarily screened according to the defect prediction threshold, and the second target region of the to-be-identified city is obtained.

[0024] Optionally, the difference defect identification of the second target region is for obtaining a target defect pipeline and a major defect of the target defect pipeline.

[0025] First water level information and second water level information of the second target region are acquired, the first water level information is used to indicate a water level condition of underground water of the second target region, and the second water level information is used to determine a water level condition in a pipeline in the second target region.

[0026] According to the first water level information, a first identification mode of the difference defect identification is determined, and a pipe network defect identification of the second target region is performed according to the first identification mode, and the target defect pipeline is obtained, the first identification mode is used to indicate a defect identification outside the pipeline of the second target region.

[0027] According to the second water level information, a second identification mode of the difference defect identification is determined, and a pipeline defect identification of the target defect pipeline is performed according to the second identification mode, and the major defect of the target defect pipeline is obtained, the second identification mode is used to indicate a defect identification inside the pipeline of the second target region.

[0028] Optionally, the method further comprises:

[0029] Pipeline information in the second target region is acquired, and a pipeline defect prediction model is called, the pipeline defect prediction model is used to predict a major defect existing in a region where the target defect pipeline is located.

[0030] The pipeline defect prediction model is controlled to perform pipeline defect prediction on the pipeline information.

[0031] According to the result of the pipeline defect prediction, a major defect of the target defect pipeline is corrected, and a correction result is re-determined as the major defect of the target defect pipeline.

[0032] Optionally, the method further comprises:

[0033] obtaining a pre-training model and a historical multi-dimensional data set of the city to be identified, the pre-training model being used to describe a development trend existing in data arranged in time sequence;

[0034] analyzing and processing the historical multi-dimensional data set to obtain a prediction variable and a target variable of the city to be identified, the prediction variable being used to indicate a factor affecting a probability or severity of a major defect of a pipe network, and the target variable being used to indicate a probability or severity of a major defect of the pipe network possibly occurring in the first to-be-identified region;

[0035] training the pre-training model according to the prediction variable and the target variable to obtain the pipe network defect prediction model.

[0036] In a second aspect, the present application provides a major defect identification device of a sewage pipe network, comprising:

[0037] an obtaining module configured to obtain a plurality of first to-be-identified regions of a city to be identified, a multi-dimensional data set, and a pipe network defect prediction model, the multi-dimensional data set being used to indicate a collection of multi-dimensional geographic information data of the city to be identified, and the first to-be-identified region comprising a plurality of second to-be-identified regions.

[0038] a processing module configured to control the pipe network defect prediction model to perform a first defect prediction on the plurality of first to-be-identified regions according to the multi-dimensional data set, to obtain a first target region, and the target region being used to indicate a region in which a major defect of a pipe network exists in the city to be identified.

[0039] The processing module is further configured to control the pipe network defect prediction model to perform a second defect prediction on a plurality of second to-be-identified regions in the first target region, to obtain a second target region of the city to be identified.

[0040] The processing module is further configured to perform a differential defect identification on the second target region, to obtain a target defect pipeline and a major defect of the target defect pipeline.

[0041] Optionally, the obtaining module is further configured to obtain first geographic information and second geographic information, and the geographic information is used to perform a multi-level division on the first to-be-identified region.

[0042] The processing module is further configured to divide the city to be identified into regions according to the first geographic information, thereby obtaining the plurality of first regions to be identified.

[0043] The processing module is further configured to divide the plurality of first regions to be identified into regions according to the second geographic information, thereby obtaining the plurality of second regions to be identified.

[0044] Optionally, the acquisition module is further configured to acquire the first region identifier of each of the plurality of first regions to be identified.

[0045] The processing module is further configured to classify the multivariate dataset according to multiple first region identifiers to obtain multiple first region datasets, wherein the region identifiers are used to distinguish the data information of the multiple regions to be identified.

[0046] The processing module is further configured to control the pipeline defect prediction model to perform defect prediction on the plurality of first regional datasets respectively, and obtain the first defect prediction value of the plurality of first regions to be identified. The defect prediction value is used to indicate the proportion of major pipeline defects in different regions of the city to be identified.

[0047] The acquisition module is further configured to acquire the defect prediction threshold of the city to be identified, the defect prediction threshold being used to indicate the lower limit of the proportion of major pipeline defects in the area to be identified.

[0048] The processing module is further configured to perform initial screening on multiple first defect prediction values ​​according to the defect prediction threshold to obtain the first target area.

[0049] Optionally, the acquisition module is further configured to acquire the second region identifiers of a plurality of second regions to be identified in the first target region.

[0050] The processing module is further configured to classify the first region dataset of the first target region according to the second region identifier to obtain a second region dataset of multiple second regions to be identified in the first target region.

[0051] The processing module is further configured to input the plurality of second region datasets into the pipeline defect prediction model to obtain second defect prediction values ​​for a plurality of second regions to be identified in the first target region.

[0052] The processing module is further configured to perform secondary filtering on multiple second defect prediction values ​​according to the defect prediction threshold to obtain the second target area of ​​the city to be identified.

[0053] Optionally, the acquisition module is further configured to acquire first water level information and second water level information of the second target area, wherein the first water level information is used to indicate the groundwater level in the second target area, and the second water level information is used to determine the water level in the pipeline in the second target area.

[0054] The major defect identification device for the sewage pipe network also includes a determination module.

[0055] The determining module is used to determine the first identification mode of the differentiated defect identification based on the first water level information.

[0056] The processing module is further configured to identify pipeline defects in the second target area according to the first identification mode to obtain the target defect pipeline, wherein the first identification mode is used to indicate the identification of defects outside the pipeline in the second target area.

[0057] The determining module is further configured to determine the second identification mode of the differentiated defect identification based on the second water level information.

[0058] The processing module is further configured to identify pipeline defects in the target defective pipeline according to the second identification mode, thereby obtaining major defects in the target defective pipeline. The second identification mode is used to indicate pipeline defect identification in the second target area.

[0059] Optionally, the acquisition module is further configured to acquire pipeline information within the second target area.

[0060] The processing module is also used to call the pipeline defect prediction model, which is used to predict major defects existing in the area where the target defect pipeline is located;

[0061] The processing module is also used to control the pipeline defect prediction model to predict pipeline defects based on the pipeline information.

[0062] The processing module is also used to correct major defects in the target defect pipeline based on the pipeline defect prediction results.

[0063] The determining module is also used to re-determine the correction result as a major defect of the target defective pipeline.

[0064] Optionally, the acquisition module is further configured to acquire a pre-trained model and a historical multivariate dataset of the city to be identified, wherein the pre-trained model is used to describe the development trends present in the data arranged in chronological order.

[0065] The processing module is also used to analyze and process historical multivariate datasets to obtain predictive variables and target variables for the city to be identified. The predictive variables are used to indicate factors that affect the probability or severity of major defects in the pipeline network, and the target variables are used to indicate the probability or severity of major defects that may occur in the pipeline network within the first area to be identified.

[0066] The processing module is further configured to train the pre-trained model based on the predicted variable and the target variable to obtain the pipeline defect prediction model.

[0067] Thirdly, this application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0068] The memory stores computer-executed instructions;

[0069] The processor executes computer execution instructions stored in the memory to implement the method for identifying major defects in sewage pipe networks as described in the first aspect and various possible implementations of the first aspect above.

[0070] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions thereon, which, when executed by a processor, are used to implement the method for identifying major defects in sewage pipe networks as described in the first aspect and various possible implementations of the first aspect.

[0071] Fifthly, this application provides a program product, including a computer program, which, when executed by a processor, implements the method for identifying major defects in sewage pipe networks as described above.

[0072] The wastewater pipe network major defect identification method, device, electronic equipment, and storage medium provided in this application acquire multiple first-to-be-identified areas of a city to be identified, a multivariate dataset, a pipe network defect prediction model, and first-area identifiers of the multiple first-to-be-identified areas. The multivariate dataset is then classified according to the multiple first-area identifiers to obtain multiple first-area datasets. The pipe network defect prediction model is controlled to predict defects in each of the multiple first-area datasets to obtain multiple first-defect prediction values. A defect prediction threshold for the city to be identified is obtained, and the multiple first-defect prediction values ​​are initially screened according to the defect prediction threshold to obtain a first target area. Second-area identifiers of multiple second-to-be-identified areas within the first target area are then acquired, and the second-area identifiers are determined according to the second-area identifiers. The method involves classifying the first target area dataset to obtain a second set of datasets for multiple second target areas. These datasets are then input into a pipeline defect prediction model to generate multiple predicted second defect values. These predicted values ​​are initially filtered according to a defect prediction threshold to identify the second target areas of the city. Based on the first and second water level information of the second target areas, a first and second identification mode for differentiated defect identification are determined. Pipeline defects in the second target areas are identified using the first identification mode to identify the target defective pipelines. Finally, pipeline defects in the target defective pipelines are identified using the second identification mode to determine the major defects of the target defective pipelines. This method not only effectively improves the accuracy and efficiency of locating and identifying major defects in sewage pipelines but also enhances the safety and reliability of the sewage pipeline system. It avoids the limitations of traditional water diversion and sealing operations, reduces the economic cost of pipeline defect detection, and enables real-time monitoring and early warning of the health status of sewage pipelines throughout their entire lifecycle, providing better protection for urban environmental protection and residents' quality of life. Attached Figure Description

[0073] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0074] Figure 1 Flowchart of the method for identifying major defects in sewage pipe networks provided in this application Figure 1 ;

[0075] Figure 2 Flowchart of the method for identifying major defects in sewage pipe networks provided in this application Figure 2 ;

[0076] Figure 3 A schematic diagram of the structure of the wastewater pipe network major defect identification device provided in this application;

[0077] Figure 4A schematic diagram of the structure of the sewage pipe network major defect identification device provided in this application.

[0078] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0079] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0080] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein.

[0081] In this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0082] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0083] With the acceleration of urbanization, the increase in urban population and the expansion of urban construction, the pressure on urban sewage treatment systems is gradually increasing, and urban sewage pipeline systems are also facing serious safety risks, making the maintenance and management of sewage pipeline systems more difficult.

[0084] Sewage pipeline systems are prone to various serious defects during long-term operation, such as pipe rupture, collapse, large-scale corrosion, and structural damage. These defects not only directly affect the sewage discharge function, but may also lead to serious public safety incidents such as environmental pollution and ground subsidence. Therefore, the demand for safety management and maintenance of sewage pipeline systems is increasing. It is necessary to establish a comprehensive pipeline information management system, strengthen preventive maintenance and regular inspections, and deal with pipeline problems in a timely manner to ensure the safe and stable operation of the system.

[0085] Currently, the technical means for inspecting sewage pipelines include closed-circuit television (CCTV) inspection, sonar inspection, pipeline periscope inspection, ground penetrating radar, etc. These technical means can be used in combination to detect and assess the condition of sewage pipelines, ensuring the normal operation and maintenance of the pipeline system.

[0086] However, existing sewage pipeline inspection technologies generally have certain limitations, as follows:

[0087] 1) Low efficiency of CCTV inspection: Although CCTV inspection can visually display the internal condition of the pipeline, it is inefficient when inspecting large areas. The inspection requires sealing and water adjustment operations, which is costly and inefficient. It is not suitable for pipeline inspection tasks that operate at high water levels and where water adjustment and sealing are difficult.

[0088] 2) Manual repeated verification is time-consuming and labor-intensive: Existing technology requires a lot of manpower to perform repeated verification, which is time-consuming and labor-intensive, and is prone to omissions, affecting the accuracy and comprehensiveness of the test results.

[0089] 3) Limited ability to prevent environmental pollution and public safety incidents: Due to the limitations of existing technology, the ability to promptly detect and repair serious defects in sewage pipeline systems is insufficient, which may lead to serious public safety incidents such as environmental pollution and ground subsidence.

[0090] The method for identifying major defects in sewage pipe networks provided in this application aims to solve the above-mentioned technical problems in the prior art.

[0091] First, the implementation scenarios involved in this application will be explained.

[0092] With the acceleration of urbanization, sewage pipe systems are prone to various serious defects during long-term operation, including corrosion, blockage, and leakage. Corrosion is caused by the erosion of pipe materials by chemicals in sewage, potentially leading to thinning, cracking, or damage to the pipe walls. Blockage occurs when solid particles, sediments, or foreign objects in the sewage clog the pipe channels, preventing normal sewage flow. Leakage can be caused by loose pipe connections, aging pipe materials, or external damage, resulting in sewage leakage, environmental pollution, and facility damage. Regular inspection and maintenance of sewage pipe systems are crucial for preventing these serious defects, ensuring the safe operation and effective drainage of the system. However, traditional inspection and maintenance methods often rely on manual patrols and periodic inspections, which are not only time-consuming and labor-intensive but also difficult to achieve comprehensive monitoring and timely early warning of the pipe system.

[0093] Therefore, this application proposes an urban sewage pipe network system equipped with a pipe network defect prediction model, which can realize intelligent monitoring and prediction of the sewage pipe system. For example, the urban sewage pipe network system equipped with the pipe network defect prediction model can acquire real-time geographic information data of the city to be identified. Based on this geographic information data, it can quickly identify and accurately locate major defects that may occur in the sewage pipes. Through big data technology, UAV remote sensing, ground penetrating radar, etc., it can acquire the health status of the urban sewage pipes throughout their entire life cycle in real time. When a sewage pipe with a major defect is identified, it can provide early warning to users, ensuring that major defects can be quickly detected and accurately located so that targeted repair measures can be taken in a timely manner, reducing the possibility of major defects in the urban sewage pipes and ensuring the stable operation of the urban sewage system and public safety.

[0094] It is understandable that the prediction of major pipeline defects in urban sewage pipe network systems in different cities is essentially the same. The following embodiments only take any one city to be identified as an example, and use the urban sewage pipe network system in the city to be identified with a pipeline defect prediction model to specifically illustrate the prediction of major pipeline defects. In addition, the pipeline defect prediction model involved in the following embodiments can identify all pipeline defects existing in the area to be identified. In the process of screening the identification results, this embodiment prioritizes major pipeline defects. The purpose of prioritizing major pipeline defects is to ensure the safety, reliability and operational efficiency of the urban sewage pipe network system, while reducing maintenance costs and improving user satisfaction. Through scientific defect prediction and priority ranking, optimal resource allocation and efficient maintenance of the urban sewage pipe network system can be achieved.

[0095] This application provides a method for identifying major defects in sewage pipe networks. It involves acquiring historical geographic information data of the urban pipe network, constructing a pipe network defect prediction model based on this data, dividing the urban pipe network into hierarchical and zoned areas to obtain multi-level pipe network regions, and acquiring real-time geographic information data of the urban pipe network based on these multi-level regions. The method then uses the pipe network defect prediction model to perform multi-level defect prediction on the real-time acquired geographic information data, accurately locating areas with major pipe network defects. For these accurately located areas, rapid detection of major pipe network defects is conducted to determine the problematic pipes and the type and severity of the defects. This method not only effectively improves the accuracy and efficiency of locating and identifying major defects in sewage pipe networks but also enhances the safety and reliability of the sewage pipe network system. It avoids the limitations of traditional water diversion and sealing operations, reduces the economic cost of pipe network defect detection, and enables real-time monitoring and early warning of the health status of sewage pipes throughout their entire life cycle, providing better protection for urban environmental protection and residents' quality of life.

[0096] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0097] Figure 1 A flowchart illustrating the method for identifying major defects in sewage pipe networks provided in this application embodiment. Figure 1 The implementing entity in this embodiment can be, for example, an urban sewage pipe network system equipped with a pipe network defect prediction model. Figure 1 As shown, the method for identifying major defects in sewage pipe networks provided in this embodiment includes:

[0098] S101: Obtain multiple first-to-be-identified areas, multivariate datasets, and pipeline defect prediction models for the city to be identified.

[0099] Among them, the multivariate dataset is used to indicate the collection of multidimensional geographic information data of the city to be identified, the first area to be identified includes multiple second areas to be identified, and the pipeline defect prediction model is used to predict the possibility of major defects in different areas of the city to be identified.

[0100] Understandably, the main entity executing this step is the urban sewage pipe network system, which stores a pipe network defect prediction model. Using this model, major defects in urban pipes can be located. To rapidly identify major defects in the sewage pipe network, the city to be identified is divided into hierarchical regions, resulting in multiple first-to-be-identified regions. The geographical scope of the city to be identified is composed of these multiple first-to-be-identified regions, meaning they are geographically connected and interconnected, forming the complete area of ​​the city. Based on these multiple first-to-be-identified regions, each region is further subdivided, resulting in multiple second-to-be-identified regions. These second-to-be-identified regions have a hierarchical relationship with their corresponding first-to-be-identified regions, and the geographical scope of each first-to-be-identified region is composed of these second-to-be-identified regions. Therefore, the hierarchical geographical division of the city to be identified is achieved, enabling the urban sewage pipe network system to accurately locate pipe defects by level and region, thus improving the efficiency of identifying and locating major pipe defects.

[0101] The multidimensional dataset includes multidimensional geographic information data of the city to be identified. This multidimensional geographic information data is a collection of data describing information such as Earth's surface features, resource distribution, and human activities. For example, it may include data on topography, climate and meteorology, water resources, population distribution, and economic development. Multidimensional geographic information data, starting from different dimensions, can assist in analyzing the overall situation and development trend of the city's sewage pipe network system. This helps to predict and locate the possible locations of major defects in the pipe network, discover potential problems and improvement opportunities, and enable urban management departments to take measures to repair and improve the network in advance. This improves the operational efficiency and management level of the pipe network and reduces the losses and impacts caused by defects.

[0102] The urban sewage pipe network system can acquire multiple first areas to be identified in the city to be identified, as well as multiple second areas to be identified in each first area. It can also acquire multi-dimensional geographic information data of the corresponding multiple areas to be identified in real time, store the multi-dimensional geographic information data in a multi-dimensional dataset, and call the pipe network defect prediction model set in the urban sewage pipe network system.

[0103] For example, four first-stage regions to be identified are pre-defined in the city to be identified: Region A, Region B, Region C, and Region D. The multivariate dataset acquired in real-time by the city's sewage pipe network system includes: the physical location, structural information, historical construction and maintenance records, and water supply data of the pipe networks in Regions A, B, C, and D; population data, economic development level, historical meteorological data, and geographic environmental information of the city to be identified. This application does not impose any special restrictions on the multivariate dataset.

[0104] Preferably, before obtaining the multivariate dataset of the city to be identified, the first area to be identified, and the pipeline defect prediction model, the method further includes:

[0105] Obtain first geographic information and second geographic information; divide the city to be identified into regions according to the first geographic information to obtain multiple first regions to be identified; divide the multiple first regions to be identified into regions according to the second geographic information to obtain multiple second regions to be identified.

[0106] Among them, geographic information is used to divide the first region to be identified into multiple levels.

[0107] Understandably, multi-level geographic information segmentation refers to gradually subdividing a large geographic region into multiple levels of sub-regions. The first and second geographic information are related, with the first geographic information at a higher level than the second, and different levels of segmentation having a subordinate relationship. Multiple different second geographic information pieces have a unique corresponding first geographic information piece. According to different hierarchical relationships, the geographic region of the city to be identified is segmented into multiple levels. That is, first, a preliminary regional segmentation is performed based on the first geographic information corresponding to the city to be identified; then, the preliminary segmentation is further subdivided according to the second geographic information corresponding to different regions, resulting in multiple levels of geographic regions. The geographic information is data pre-stored in the urban sewage pipe network system. When the urban sewage pipe network system performs regional segmentation of the city to be identified, it can call upon the corresponding geographic information to achieve multi-level segmentation of the city to be identified.

[0108] The system acquires the first and second geographic information stored in the city's sewage pipe network system, with the first geographic information at a higher level than the second geographic information. That is, the regional division based on the second geographic information is achieved on the basis of the regional division based on the first geographic information. First, the city to be identified is divided into regions according to the first geographic information, resulting in multiple sub-regions at corresponding levels, which are also multiple first regions to be identified. The second geographic information corresponding to each of the different first regions to be identified is determined, and the multiple first regions to be identified are divided into regions according to the determined second geographic information, resulting in multiple sub-regions after the division of the first regions to be identified, and these sub-regions are identified as second regions to be identified.

[0109] For example, the first geographic information can be the district / county-level administrative boundary of the city to be identified. The sewage pipe network area of ​​the city to be identified is divided according to this district / county-level administrative boundary, and the division result is determined as the first area to be identified. The second geographic information can be the service range of multiple pumping stations in different first areas to be identified. The corresponding first areas to be identified are further divided according to the service range of different pumping stations in different first areas to be identified, and the result of this division according to the service range of the pumping stations is determined as the second area to be identified. In other words, the geographic area of ​​the city to be identified is first initially divided according to the administrative district / county boundaries of the city to be identified, and then further divided according to the service range of pumping stations within different administrative districts / counties, thus obtaining multiple first areas to be identified and multiple second areas to be identified. This application does not impose any special restrictions on the type of geographic information.

[0110] Preferably, the method further includes:

[0111] Obtain the pre-trained model and the historical multivariate dataset of the city to be identified; analyze and process the historical multivariate dataset to obtain the predictor variables and target variables of the city to be identified; train the pre-trained model based on the predictor variables and target variables to obtain the pipeline defect prediction model.

[0112] Among them, the pre-trained model is used to describe the development trend in the data arranged in chronological order, the predictor variable is used to indicate the factors that affect the probability or severity of major defects in the pipeline network, and the target variable is used to indicate the probability or severity of major defects that may occur in the pipeline network in the first area to be identified.

[0113] Understandably, a multivariate dataset includes multidimensional geographic information data of the city to be identified, while a historical multivariate dataset shows the changes in geographic information data of the city to be identified within a corresponding historical period. Using the historical multivariate dataset, the causes and influencing factors of major pipeline defects in the sewage pipeline system of the city to be identified within a historical period can be analyzed. Before constructing a pipeline defect prediction model using historical multivariate data, it is necessary to integrate and process the historical multivariate dataset to determine the effective data information, that is, the data information related to the occurrence of major pipeline defects, and to filter out the invalid data information, thereby improving the accuracy of the pipeline defect prediction model.

[0114] A pre-trained model and historical multivariate datasets of the city to be identified are obtained. These datasets are then analyzed, removing data irrelevant to the occurrence of major pipeline defects. Further filtering is performed based on factors influencing the probability or severity of major pipeline defects, yielding the input information for the pre-trained model, i.e., the predictor variables. Under different factors influencing major pipeline defects, the probability or severity of potential major pipeline defects under different conditions is determined, yielding the output information for the pre-trained model, i.e., the target variable. The pre-trained model is then trained using the determined predictor and target variables to obtain a training model capable of predicting the probability of major pipeline defects in the area to be identified. This training model is then designated as the pipeline defect prediction model.

[0115] For example, a pre-trained model can be a time series model. It integrates various types of big data from historical multivariate datasets to construct a multi-dimensional database containing information such as the physical location and structure of the pipeline network, historical construction and maintenance records, water supply data, population data, economic development level, historical meteorological data, and geographic environmental information. This multi-dimensional database is then cleaned to remove invalid, erroneous, and duplicate data, and missing data is filled in to ensure data quality and consistency. Based on the cleaned data, a prediction model is built, utilizing machine learning, deep learning, and other technologies. Factors such as pipeline network data, economic development, meteorology, water supply, rainfall, population, and geographic information are used as input variables, and the output variable is the probability or severity of potential pipeline network defects within the target area. This allows the time series model to predict potential pipeline network defect areas, as well as the corresponding defect types and probabilities, based on historical trends in various data types. The trained time series model is then identified as a pipeline network defect prediction model.

[0116] S102: Control the pipeline defect prediction model to perform initial defect prediction on the multiple first areas to be identified according to the multivariate dataset, and obtain the first target area.

[0117] S103: Control the pipeline defect prediction model to perform secondary defect prediction on multiple second areas to be identified in the first target area to obtain the second target area of ​​the city to be identified.

[0118] The target area is used to indicate the area in the city to be identified that has major pipeline defects; the first target area is used to indicate the area with the highest probability of having major pipeline defects among multiple first areas to be identified; and the second target area is used to indicate the area with the highest probability of having major pipeline defects among multiple second areas to be identified corresponding to the first target area.

[0119] Understandably, pipeline defect prediction models can predict the probability of major pipeline defects in different areas to be identified, and then select the areas with the highest probability of major pipeline defects. By predicting major pipeline defects in multiple levels of areas to be identified, the areas where major pipeline defects may exist are gradually narrowed down, and finally the specific defect location is located, thereby improving the accuracy and efficiency of pipeline defect prediction and location.

[0120] The multivariate dataset of the city to be identified is input into the pipeline defect prediction model. The pipeline defect prediction model can classify the multivariate dataset according to the division of the first area to be identified, so as to make pipeline defect prediction for different first areas to be identified. The pipeline defect prediction model is controlled to make defect prediction on the classified multivariate dataset, that is, to make initial defect prediction for multiple first areas to be identified. According to the prediction results, the probability of the existence of major pipeline defects in multiple first areas to be identified in the city to be identified can be determined, that is, the probability of the existence of major pipeline defects in multiple first areas to be identified is obtained. The probabilities of the multiple major pipeline defects are filtered to determine the first area to be identified with the highest probability, and the first area to be identified with the highest probability is determined as the first target area.

[0121] Since the regional division of the city to be identified is multi-level, and the second region to be identified is obtained by further refining the first region to be identified, after determining the first target region of the city to be identified, in order to more accurately locate the possible major pipeline defects in the city to be identified, the pipeline defect prediction model is controlled to perform secondary defect prediction on multiple second regions to be identified in the currently determined first target region. According to the prediction results, the probability of major pipeline defects existing in multiple second regions to be identified in the first target region can be determined, that is, the probability of major pipeline defects existing in multiple second regions to be identified can be obtained. The probabilities of major pipeline defects are then filtered to determine the second region to be identified with the highest probability, and this second region to be identified with the highest probability is determined as the second target region.

[0122] Understandably, the first target area and the second target area are related. The first target area is the area with the highest probability of having major pipeline defects, which is obtained from multiple first areas to be identified. The second target area is based on the first target area and further determines the specific location information of the major pipeline defects within the range of the first target area.

[0123] For example, the cities to be identified include: first target area A, first target area B, first target area C, and first target area D. The multivariate data of the cities to be identified obtained in this instance are input into a pipeline defect prediction model. This model predicts the probability of major pipeline defects existing in multiple first target areas. The resulting defect probabilities for different first target areas are: first target area A - probability A, first target area B - probability B, first target area C - probability C, and first target area D - probability D. The order of these probabilities is: probability B > probability A > probability D > probability C. Therefore, the area with the highest probability of major pipeline defects is first target area B, and this first target area B is identified as the first target area. The pipeline defect prediction model is then controlled to predict the probability of major pipeline defects in first target area B. Defect prediction is performed on the data information corresponding to multiple second-to-be-identified areas to achieve secondary defect prediction for multiple second-to-be-identified areas in the first-to-be-identified area B. The multiple second-to-be-identified areas corresponding to the first-to-be-identified area B are: second-to-be-identified area B1, second-to-be-identified area B2, second-to-be-identified area B3, and second-to-be-identified area B4. The defect probabilities of the obtained second-to-be-identified areas are: second-to-be-identified area B1 - probability B1, second-to-be-identified area B2 - probability B2, second-to-be-identified area B3 - probability B3, and second-to-be-identified area B4 - probability B4. The order of the probabilities is: probability B3 > probability B4 > probability B2 > probability B1. Therefore, after precise positioning, the area with the highest probability of having a major pipeline defect is the second-to-be-identified area B3, and this second-to-be-identified area B3 is determined as the second target area.

[0124] S104: Perform differential defect identification on the second target area to obtain the target defect pipe and the major defects of the target defect pipe.

[0125] Among them, the target defect pipeline is used to indicate the pipeline in the urban sewage pipeline network system where there are major pipeline defects. Major defects include: defect type and defect severity.

[0126] Understandably, major defects in urban sewage pipe networks refer to defects in the sewage pipe itself or its surrounding area that affect the strength, rigidity, and stability of the network structure, and will cause serious environmental pollution or pose significant safety hazards. Major defects in pipe networks can be categorized into two types based on the issues involved: one type involves damage and deformation of the pipe material, and the other involves the support and stability of the surrounding soil. Specifically, defects in the pipe itself mainly include damage and deformation of the pipe material, while cavities in the surrounding soil mainly involve soil support and stability issues. Therefore, the types of major defects in pipe networks include: defects in the pipe itself and cavities in the surrounding soil.

[0127] In urban sewage pipe network systems, the severity of pipe network defects and their impact on pipeline operation are important bases for assessing and classifying pipe network defects. Based on their impact on the overall pipe structure and operation, their potential to cause sewage pipe failure, and the resulting safety hazards, major pipe network defects can be classified into four levels. For example, defects at levels 1-2 are generally minor or moderate, with minimal impact on pipeline operation, but require regular monitoring and maintenance. Defects at levels 3-4 are severe or extremely severe, with a significant impact on pipeline operation and safety, requiring immediate repair or replacement to ensure safe pipeline operation.

[0128] Acquire multi-dimensional geographic information data within the current second target area, analyze and process this multi-dimensional geographic information data, determine the applicable defect identification method for the current second target area, and adopt a defect identification method that matches the current geographic information data, which can effectively improve the accuracy and efficiency of defect detection; perform differentiated defect identification on the current second target area according to the defect identification method, obtain the target defect pipeline and the major defects of the target defect pipeline, as well as the defect type and defect degree of the major defects.

[0129] For example, for the high-risk area identified as potentially having major pipeline defects, namely the second area to be identified (B3), rapid detection of major pipeline defects is carried out. Since different pipeline defect identification methods are applicable to different environments and conditions, each with its own unique advantages and limitations, selecting the appropriate detection method requires comprehensive consideration of the pipeline's material, shape, environmental conditions, and detection requirements to achieve the best detection results. Existing technologies for detecting sewage pipelines include CCTV inspection, sonar inspection, pipeline periscope inspection, and ground-penetrating radar. CCTV inspection is suitable for scenarios requiring detailed and intuitive understanding of the pipeline's internal condition, especially for pipelines with large diameters that are easily accessible to camera equipment. Sonar inspection is suitable for pipelines with complex internal environments or large amounts of sediment. Ground-penetrating radar is suitable for buried pipelines, especially plastic or concrete pipelines. For the identified problematic pipelines, a combination of multiple advanced geophysical exploration technologies is used to conduct rapid internal pipeline inspection, thereby accurately obtaining the type and severity of defects in the problematic pipelines.

[0130] The method for identifying major defects in sewage pipe networks provided in this embodiment acquires multiple first-to-be-identified areas of the city to be identified, a multivariate dataset, and a pipe network defect prediction model. The multivariate dataset is input into the pipe network defect prediction model, which then performs initial defect prediction on the multiple first-to-be-identified areas to obtain a first target area. Next, the model performs secondary defect prediction on multiple second-to-be-identified areas within the obtained first target area to obtain a second target area of ​​the city to be identified. Differential defect identification is then performed on the second target area to obtain the target defective pipe and its major defects. This method not only effectively improves the accuracy and efficiency of locating major defects in sewage pipe networks but also enhances the safety and reliability of the sewage pipe network system, enabling real-time monitoring of the health status of sewage pipes throughout their entire lifecycle.

[0131] Figure 2 A flowchart illustrating the method for identifying major defects in sewage pipe networks provided in this application embodiment. Figure 2 .like Figure 1 As shown, in this embodiment... Figure 3 Based on the embodiments, the method for identifying major defects in sewage pipe networks is described in detail. The method for identifying major defects in sewage pipe networks shown in this embodiment includes:

[0132] S201: Obtain multiple first-to-be-identified areas, multivariate datasets, and pipeline defect prediction models for the city to be identified.

[0133] Step S201 is similar to step S101 above, and will not be repeated here.

[0134] S202: Obtain the first region identifiers of the plurality of first regions to be identified, and classify the multivariate dataset according to the plurality of first region identifiers to obtain a plurality of first region datasets.

[0135] S203: Control the pipeline defect prediction model to perform defect prediction on the multiple first region datasets respectively, and obtain the first defect prediction value of the multiple first regions to be identified.

[0136] Among them, the area identifier is used to distinguish the data information of multiple areas to be identified, the area dataset is used to indicate the collection of multi-dimensional geographic information data of the areas to be identified, and the defect prediction value is used to indicate the proportion of major pipeline defects in different areas of the city to be identified.

[0137] Understandably, a region identifier is a unique identifier used to distinguish and identify different geographical regions. For example, it can be a number, code, name, or other form of identifier used to clearly distinguish different regions. In order to conduct rapid identification of major defects in sewage pipe networks, the cities to be identified were divided into hierarchical regions, resulting in multiple first regions to be identified. Each first region to be identified has a unique corresponding region identifier, thereby distinguishing different first regions to be identified. A regional dataset is a collection of multi-dimensional geographic information data containing data within a specific region. It can include various attributes and information of the region, such as geographic coordinates, area, population density, infrastructure information, etc.

[0138] A multivariate dataset is a collection of comprehensive multivariate geographic information data obtained for a city to be identified. When identifying defects in a city, it can identify defects in multiple first-to-be-identified areas, determine the proportion of major pipeline defects in different first-to-be-identified areas in the city, and thus more accurately locate areas in the city that may have major pipeline defects.

[0139] First region identifiers of multiple first regions to be identified are obtained respectively, and the multivariate datasets are classified according to the multiple first region identifiers. The changes of multidimensional geographic information data in multiple first regions to be identified are determined respectively, and the data sets of different first regions to be identified in this time are determined as the corresponding first region datasets. The pipeline network defect prediction model performs defect prediction on multiple first region datasets respectively, obtains the proportion of major pipeline network defects in multiple first regions to be identified, and determines the defect proportion of different first regions to be identified as the first defect prediction value of the corresponding first region to be identified.

[0140] Understandably, urban sewage pipe networks play a crucial role in urban infrastructure, primarily functioning to effectively collect and discharge sewage, prevent floods and drain water, protect water resources, safeguard public health, and improve the ecological environment, thereby ensuring the normal operation and sustainable development of cities. However, various defects may exist in urban sewage pipe networks, which could significantly impact the system's operation and safety. By using a pipe network defect prediction model to predict defects in multiple first-to-be-identified areas, the proportion of defects in each first-to-be-identified area is obtained, i.e., the defect prediction value. This defect prediction value can reflect the predicted situation of major pipe network defects that may exist in the first-to-be-identified areas of the city to be identified, specifically including the probability of different defect types and the degree of defects corresponding to different defect types.

[0141] For example, four first areas to be identified are pre-set in the city to be identified: first area to be identified A, first area to be identified B, first area to be identified C, and first area to be identified D. The first area identifiers of the corresponding first areas to be identified are obtained, and these first area identifiers are area identifier A, area identifier B, area identifier C, and area identifier D, respectively. The multivariate dataset is classified according to the multiple first area identifiers to obtain multiple first area datasets corresponding to the first areas to be identified A, B, C, and D, respectively. Then, the pipeline defect prediction model is controlled to perform defect prediction on the multiple first area datasets to obtain the first defect prediction values ​​of the multiple first areas to be identified, and the corresponding first defect prediction values ​​are defect prediction value A, defect prediction value B, defect prediction value C, and defect prediction value D, respectively.

[0142] S204: Obtain the defect prediction threshold of the city to be identified, and perform initial screening on multiple first defect prediction values ​​according to the defect prediction threshold to obtain the first target area.

[0143] The defect prediction threshold is used to indicate the lower limit of the proportion of major pipeline defects in the area to be identified.

[0144] Understandably, the pipeline defect prediction model can output defect prediction values ​​for major pipeline defects within multiple first-to-be-identified areas, and these predicted values ​​vary. Therefore, a threshold is needed to determine whether a major pipeline defect exists in a specific first-to-be-identified area; this is the defect prediction threshold. The urban sewage pipeline system pre-stores this defect prediction threshold as a reference value to measure the presence of major pipeline defects within the first-to-be-identified area. This threshold represents the lower limit of the proportion of major pipeline defects in the corresponding first-to-be-identified area. If the defect prediction value of a first-to-be-identified area is higher than the defect prediction threshold, it can be determined that a major pipeline defect exists in that area. If the defect prediction value of a first-to-be-identified area is lower than the defect prediction threshold, it can be determined that no major pipeline defect exists in that area. By setting a defect prediction threshold, areas requiring focused attention and maintenance—i.e., high-risk areas—can be effectively distinguished among multiple first-to-be-identified areas. Thus, based on the proportion of major pipeline defects in different first-to-be-identified areas, large areas can be identified as smaller areas, improving the efficiency of the urban sewage pipeline system in locating major pipeline defects. This application does not impose any special restrictions on the determination of the defect prediction threshold.

[0145] The urban sewage pipe network system can obtain the defect prediction threshold of the city to be identified in the current time, and compare the defect prediction threshold with the first defect prediction value of multiple first defect prediction values ​​to achieve the initial screening of multiple first defect prediction values; according to the comparison results, multiple first defect prediction values ​​that are greater than or equal to the defect prediction threshold are selected, and the multiple first defect prediction values ​​corresponding to the multiple first defect prediction values ​​obtained in the current screening are determined as the first target area.

[0146] Understandably, the number of first target regions can be multiple or one, and the number of first target regions is determined by the comparison results of multiple first defect prediction values ​​and defect prediction thresholds in the current instance.

[0147] For example, four first areas to be identified are pre-set in the city to be identified: first area to be identified A, first area to be identified B, first area to be identified C, and first area to be identified D, and the corresponding first defect prediction values ​​are: defect prediction value A, defect prediction value B, defect prediction value C, and defect prediction value D, respectively. The defect prediction threshold for the city to be identified can be the defect prediction value E, and the defect prediction value E is compared with multiple first defect prediction values. If the defect prediction values ​​A, C, and D are all less than the defect prediction value E, it indicates that according to the judgment criteria of the urban sewage pipe network system, there are no major pipe network defects in the first areas to be identified A, C, and D. If the defect prediction value B is greater than the defect prediction value E, it indicates that according to the judgment criteria of the urban sewage pipe network system, there are major pipe network defects in the first area to be identified B, and the first area to be identified B is determined as the first target area, thereby performing more detailed defect identification in the first area to be identified B.

[0148] S205: Obtain the second region identifiers of multiple second regions to be identified in the first target region respectively.

[0149] S206: Classify the first region dataset of the first target region according to the second region identifier to obtain a second region dataset of multiple second regions to be identified in the first target region.

[0150] S207: Input the multiple second region datasets into the pipeline defect prediction model to obtain the second defect prediction values ​​of multiple second regions to be identified in the first target region.

[0151] S208: Perform secondary filtering on multiple second defect prediction values ​​according to the defect prediction threshold to obtain the second target area of ​​the city to be identified.

[0152] Understandably, the first and second region identifiers are related; both can distinguish data information from different regions to be identified. The only difference between the two identifiers is at the region level, allowing them to differentiate data information from different levels of regions to be identified. Since the regional division of the cities to be identified is done in a hierarchical manner, their corresponding geographical areas are first divided into multiple first regions to be identified. Then, each of these first regions is further subdivided to obtain multiple second regions to be identified. Therefore, each first region to be identified has a unique corresponding geographical area composed of multiple second regions to be identified. Thus, the first region dataset corresponding to the first region to be identified is a collection of multi-dimensional geographic information data from multiple second regions to be identified. Based on the multiple second region identifiers corresponding to the first region to be identified, the first region dataset can be divided into multiple second region datasets corresponding to the second regions to be identified.

[0153] The second region identifiers of multiple second regions to be identified in the first target area are obtained respectively. The first region dataset corresponding to the first target area is classified according to the multiple second region identifiers obtained in this time, and the first region dataset is divided into multiple data sets corresponding to multiple second regions to be identified, that is, the second region dataset of multiple second regions to be identified in the first target area is obtained. The multiple second region datasets are input into the pipeline network defect prediction model, and the pipeline network defect prediction model is controlled to identify defects in multiple second regions to be identified, and the second defect prediction value of multiple second regions to be identified in the first target area is obtained. Then, the defect prediction threshold is compared with the second defect prediction value of multiple second regions to be identified, and a secondary filtering process is carried out on multiple second defect prediction values. According to the comparison results, multiple second defect prediction values ​​that are greater than or equal to the defect prediction threshold are filtered out, and the multiple second regions to be identified corresponding to the multiple second defect prediction values ​​obtained in this filtering are determined as the second target areas.

[0154] For example, if the first target area determined in this instance is: the first area to be identified, B, and the multiple second areas to be identified corresponding to the first area to be identified, B1, B2, B3, and B4, respectively, with corresponding area identifiers B1, B2, B3, and B4, respectively; the area dataset of the first area to be identified, B, is classified according to the multiple second area identifiers to obtain multiple second area datasets corresponding to the second areas B1, B2, B3, and B4, respectively; then the pipeline defect prediction model is controlled to perform defect prediction on the multiple second area datasets to obtain the second defect prediction values ​​for the multiple second areas to be identified, with the corresponding second defect prediction values ​​being: defect prediction value B1, defect prediction value B2, defect prediction value B3, and defect prediction value B4; since the defect prediction values ​​can reflect the proportion of major pipeline defects in different areas of the city to be identified, therefore... Therefore, the multiple predicted values ​​of the second defects determined in this instance can be converted into probability values, and the defect prediction threshold can also be converted into a corresponding probability value, with the probability value corresponding to the defect prediction threshold being probability E. The defect probabilities of the obtained second areas to be identified are as follows: second area to be identified B1 - probability B1, second area to be identified B2 - probability B2, second area to be identified B3 - probability B3, second area to be identified B4 - probability B4, and the order of the probabilities is: probability B3 > probability B4 > probability B2 > probability B1. The probability values ​​corresponding to the multiple predicted values ​​of the second defects are compared with probability E. If probability B1, probability B2, and probability B4 are all less than probability E, it indicates that according to the judgment criteria of the urban sewage pipe network system, there are no major pipe network defects in the second areas to be identified B1, B2, and B4. If probability B3 is greater than probability E, it indicates that according to the judgment criteria of the urban sewage pipe network system, there are major pipe network defects in the second area to be identified B3, and the second area to be identified B3 is determined as the second target area.

[0155] It is understood that the determination of the target area in the above embodiments can be based on the area with the highest probability of having a major pipeline defect among multiple areas to be identified, or it can be based on a pre-set defect prediction threshold, with the area corresponding to the defect prediction value that is greater than or equal to the threshold being used as the target area. This application does not impose any special restrictions on the determination of the target area.

[0156] S209: Obtain the first water level information and the second water level information of the second target area.

[0157] S210: Based on the first water level information, determine the first identification mode for differentiated defect identification, and perform pipeline defect identification on the second target area according to the first identification mode to obtain the target defect pipeline.

[0158] S211: Based on the second water level information, determine the second identification mode for differentiated defect identification, and identify the pipeline defect of the target defect pipeline according to the second identification mode to obtain the major defects of the target defect pipeline.

[0159] The first water level information is used to indicate the groundwater level in the second target area, the second water level information is used to determine the water level inside the pipeline in the second target area, the first identification mode is used to indicate the identification of defects outside the pipeline in the second target area, and the second identification mode is used to indicate the identification of defects inside the pipeline in the second target area.

[0160] Understandably, the first identification mode is the method for identifying external defects in the pipe network that is determined based on the type of leakage problem in the second target area. The second identification mode is the method for identifying internal defects in the pipe network that is determined after the target defective pipe is identified, based on whether the water level in the pipe is full. According to the type of major defects in the pipe network, the urban sewage pipe network system can use different defect detection methods. After determining the target area where major defects in the pipe network may exist, the urban sewage pipe network system performs differentiated identification on the second target area. First, it identifies the pipes in the second target area that may have major defects in the pipe network, and then performs defect identification on the pipes to determine the type and extent of the major defects currently existing in the pipes.

[0161] Based on the first water level information of the second target area, the type of leakage problem in the pipeline network within the second target area can be determined. Specifically, the leakage problem includes: external sewage seepage and internal seepage of external water. After identifying the pipeline with a major pipeline defect, based on the second water level information of the second target area, it can be determined whether the water level in the pipeline is full. Specifically, the water level in the pipeline can be divided into full water level and non-full water level. The defect identification methods and equipment selection are different in the full water level and non-full water level states.

[0162] For example, when the groundwater level is higher than the pipe burial depth, major defects in the pipe body will cause external water to seep into the pipe, i.e., external water seepage into the pipe. In this case, a combination of water quality characteristic factor technology and hydrodynamic model inversion technology can be used to identify sewage pipe sections that may have major defects. When the groundwater level is lower than the pipe burial depth, major defects in the pipe body will cause sewage to seep out. For sewage seepage, a UAV equipped with radar and infrared thermal imager can be used to quickly detect leaks through remote sensing technology. Depending on the water level inside the pipe, different combinations of geophysical exploration technologies can be used. Specifically, for pipes operating at full water level, without the need for sealing, water diversion, or dredging, the focused current method can be combined with sonar technology to detect pipe defects. For pipes not operating at full water level, a comprehensive detection method combining CCTV technology, focused current method, sonar technology, and in-pipe radar technology can be used. Among them, CCTV technology is used to monitor the defects of the pipe above the water surface, the focused current method is used to detect the leakage location of the pipe below the water surface, sonar technology is used to detect circumferential defects of the pipe, and in-pipe radar is used to detect cavities in the soil around the pipe.

[0163] The system acquires first and second water level information for the second target area. Based on the first water level information acquired in the second target area, it compares the groundwater level with the pipeline burial depth to determine the type of leakage problem in the pipeline network within the second target area. Based on the current leakage situation of the pipeline network, it determines a first identification mode for differentiated defect identification and performs pipeline network defect identification in the second target area according to the first identification mode to obtain the target defective pipeline. Based on the second water level information, it determines the current water level status of the target defective pipeline, i.e., whether the water level status of the target defective pipeline is full or not. Based on different water level statuses, it determines a second identification mode for differentiated defect identification and performs pipeline defect identification within the target defective pipeline according to the second identification mode to obtain the major defects of the target defective pipeline, as well as the defect type and degree of the major defects.

[0164] For example, if the first water level information obtained is that the groundwater level is higher than the pipe burial depth, it indicates that the leakage problem in the second area B3 to be identified is: external water seepage into the interior. For this problem, a combination of water quality characteristic factor technology and hydrodynamic model inversion technology can be used as the first identification mode for differentiated defect identification. The pipeline network defect identification in the second area B3 to be identified can be performed according to the first identification mode to obtain the target defective pipeline. If the second water level information obtained is that the water level inside the target defective pipeline is at full water level, it indicates that the target defective pipeline is a pipeline operating at full water level. The focused current method can be combined with sonar technology to identify the defect in the target defective pipeline, thereby accurately obtaining the target defective pipeline and the defect type and defect level existing in the pipe at this time. The defect type can be a pipeline body defect, and the defect level can be level 4.

[0165] Optionally, the method further includes:

[0166] Obtain pipeline information within the second target area and invoke the pipeline defect prediction model; control the pipeline defect prediction model to predict pipeline defects based on the pipeline information; based on the pipeline defect prediction results, correct the major defects of the target defect pipeline and re-determine the corrected results as the major defects of the target defect pipeline.

[0167] The pipeline information may include, for example, the pipeline burial depth, pipeline length, pipeline condition, surrounding soil condition, equivalent diameter of the pipeline, and pipeline material; the pipeline defect prediction model is used to predict major defects existing in the area where the target defective pipeline is located.

[0168] Understandably, when underground pipelines leak, the water flow may erode the soil around the pipeline. Over time, the water flow will carry away fine soil particles, gradually forming cavities. Therefore, when identifying the target defective pipeline and its corresponding major defects based on leakage problems in the second target area, it is also necessary to consider the cavitation problem in the surrounding soil caused by the leakage. A pipeline defect prediction model is used to identify whether the target defective pipeline has cavities in the surrounding soil, thereby improving the accuracy of the defect identification results obtained from the urban sewage pipeline system. The pipeline defect prediction model is mainly trained on the cavitation problem in the surrounding soil of the pipeline. The resulting prediction model can effectively predict the cavitation situation in the soil around the pipeline. Through analysis and training on a large amount of pipeline inspection data, the model can identify potential soil cavities and assess their potential impact on the integrity and safety of the pipeline structure.

[0169] The system acquires pipeline information within the second target area and invokes a pipeline defect prediction model. The pipeline information is input into the model, which then predicts pipeline defects in the target pipeline to determine if there are cavities in the surrounding soil. Based on the prediction results, major defects in the target pipeline are corrected, and the corrected results are redefined as the defect identification results for the target pipeline. This update of the major defect results improves the accuracy of major defect identification in the urban sewage network system.

[0170] For example, a combination of logistic regression prediction models and ground-based vehicle-mounted ground-penetrating radar (GPR) can be used for rapid detection of soil cavities around pipes. First, a logistic regression model is constructed to predict pipe networks prone to collapse. Then, GPR is used to quickly detect high-risk pipe networks, thereby determining whether the target defective pipe has soil cavities around it. Based on the identification results, the defect types and severity of major defects identified through differentiated defect identification are supplemented and improved. Specifically, if soil cavities around the target defective pipe are identified, and the differentiated defect identification result is: the defect type within the target defective pipe is a pipe body defect, then... If the defect level is 4, and the differential defect identification result is corrected, then the corrected defect identification result can be: the defect type in the target defective pipeline is a pipeline body defect and a cavity in the surrounding soil, with a defect level of 4. If the target defective pipeline is found to have no cavity in the surrounding soil, and the differential defect identification result is: the defect type in the target defective pipeline is a pipeline body defect, with a defect level of 4, then no correction is needed for the differential defect identification result. The identification result for the major pipeline network defect can be: the defect type in the target defective pipeline is a pipeline body defect, with a defect level of 4. This application does not impose any special restrictions on the form of correction processing for the differential defect identification results.

[0171] The wastewater pipe network major defect identification method provided in this embodiment obtains multiple first-to-be-identified areas of the city to be identified, a multivariate dataset, a pipe network defect prediction model, and first-area identifiers for the multiple first-to-be-identified areas. The multivariate dataset is then classified according to these first-area identifiers to obtain multiple first-area datasets. The pipe network defect prediction model is then used to predict defects in each of the multiple first-area datasets to obtain first-defect prediction values ​​for the multiple first-to-be-identified areas. A defect prediction threshold for the city to be identified is then obtained, and the multiple first-defect prediction values ​​are initially screened according to the defect prediction threshold to obtain a first target area. Finally, second-area identifiers for multiple second-to-be-identified areas within the first target area are obtained, and the resulting first target area is classified according to the second-area identifiers. The first region dataset of the target area is classified to obtain a second region dataset of multiple second regions to be identified within the first target area. These second region datasets are then input into a pipeline defect prediction model to obtain predicted values ​​for second defects in the multiple second regions to be identified within the first target area. These predicted values ​​are then initially filtered according to a defect prediction threshold to obtain the second target areas of the city to be identified. First and second water level information of the second target areas are acquired, and first and second identification modes for differentiated defect identification are determined. Pipeline defects in the second target areas are identified according to the first identification mode to obtain the target defective pipelines. Then, pipeline defects in the target defective pipelines are identified according to the second identification mode to obtain the major defects of the target defective pipelines. This method not only effectively improves the accuracy and efficiency of locating and identifying major defects in sewage pipelines, but also enhances the safety and reliability of the sewage pipeline system. It avoids the limitations of traditional water diversion and sealing operations, reduces the economic cost of pipeline defect detection, and enables real-time monitoring and early warning of the health status of sewage pipelines throughout their entire life cycle, providing better protection for urban environmental protection and residents' quality of life.

[0172] Figure 3 A schematic diagram of the structure of the wastewater pipe network major defect identification device provided in this application. Figure 4 As shown, this application provides a major defect identification device for sewage pipe networks. The major defect identification device 300 for sewage pipe networks includes:

[0173] The acquisition module 301 is used to acquire multiple first areas to be identified, a multivariate dataset, and a pipeline defect prediction model for the city to be identified. The multivariate dataset is used to indicate a collection of multidimensional geographic information data of the city to be identified. The first areas to be identified include multiple second areas to be identified.

[0174] Processing module 302 is used to control the pipeline defect prediction model to perform initial defect prediction on the multiple first areas to be identified according to the multivariate dataset, and obtain a first target area, wherein the target area is used to indicate the area in the city to be identified where there are major pipeline defects.

[0175] The processing module 302 is also used to control the pipeline defect prediction model to perform secondary defect prediction on multiple second areas to be identified in the first target area, so as to obtain the second target area of ​​the city to be identified.

[0176] The processing module 302 is also used to perform differentiated defect identification on the second target area to obtain the target defect pipe and the major defects of the target defect pipe.

[0177] Optionally, the acquisition module 301 is further configured to acquire first geographic information and second geographic information, wherein the geographic information is used to perform multi-level division of the first area to be identified.

[0178] The processing module 302 is further configured to divide the city to be identified into regions according to the first geographic information to obtain the plurality of first regions to be identified.

[0179] The processing module 302 is further configured to divide the plurality of first regions to be identified into regions according to the second geographic information, thereby obtaining the plurality of second regions to be identified.

[0180] Optionally, the acquisition module 301 is further configured to acquire the first region identifier of each of the plurality of first regions to be identified.

[0181] The processing module 302 is further configured to classify the multivariate dataset according to multiple first region identifiers to obtain multiple first region datasets, wherein the region identifiers are used to distinguish the data information of the multiple regions to be identified.

[0182] The processing module 302 is further configured to control the pipeline defect prediction model to perform defect prediction on the plurality of first regional datasets respectively, and obtain the first defect prediction value of the plurality of first regions to be identified. The defect prediction value is used to indicate the proportion of major pipeline defects in different regions of the city to be identified.

[0183] The acquisition module 301 is further configured to acquire the defect prediction threshold of the city to be identified, the defect prediction threshold being used to indicate the lower limit of the proportion of major pipeline defects in the area to be identified.

[0184] The processing module 302 is further configured to perform initial screening of multiple first defect prediction values ​​according to the defect prediction threshold to obtain the first target area.

[0185] Optionally, the acquisition module 301 is further configured to acquire the second region identifiers of a plurality of second regions to be identified in the first target region.

[0186] The processing module 302 is further configured to classify the first region dataset of the first target region according to the second region identifier to obtain a second region dataset of multiple second regions to be identified in the first target region.

[0187] The processing module 302 is further configured to input the plurality of second region datasets into the pipeline defect prediction model to obtain second defect prediction values ​​for a plurality of second regions to be identified in the first target region.

[0188] The processing module 302 is further configured to perform secondary filtering on multiple second defect prediction values ​​according to the defect prediction threshold to obtain the second target area of ​​the city to be identified.

[0189] Optionally, the acquisition module 301 is further configured to acquire first water level information and second water level information of the second target area, wherein the first water level information is used to indicate the groundwater level in the second target area, and the second water level information is used to determine the water level in the pipeline in the second target area.

[0190] The major defect identification device for the sewage pipe network also includes: a determination module 303.

[0191] The determining module 303 is used to determine the first identification mode of the differentiated defect identification based on the first water level information.

[0192] The processing module 302 is further configured to identify pipeline defects in the second target area according to the first identification mode to obtain the target defect pipeline, wherein the first identification mode is used to indicate the identification of defects outside the pipeline in the second target area.

[0193] The determining module 303 is further configured to determine the second identification mode of the differentiated defect identification based on the second water level information.

[0194] The processing module 302 is further configured to identify pipeline defects in the target defect pipeline according to the second identification mode, thereby obtaining major defects in the target defect pipeline. The second identification mode is used to indicate pipeline defect identification in the second target area.

[0195] Optionally, the acquisition module 301 is further configured to acquire pipeline information within the second target area.

[0196] The processing module 302 is also used to call the pipeline defect prediction model, which is used to predict major defects existing in the area where the target defect pipeline is located;

[0197] The processing module 302 is also used to control the pipeline defect prediction model to predict pipeline defects based on the pipeline information.

[0198] The processing module 302 is further configured to correct major defects in the target defect pipeline based on the results of the pipeline defect prediction.

[0199] The determining module 303 is further configured to re-determine the correction result as a major defect of the target defective pipeline.

[0200] Optionally, the acquisition module 301 is further configured to acquire a pre-trained model and a historical multivariate dataset of the city to be identified, wherein the pre-trained model is used to describe the development trend present in the data arranged in chronological order.

[0201] The processing module 302 is further configured to analyze and process the historical multivariate dataset to obtain predictive variables and target variables for the city to be identified. The predictive variables are used to indicate factors that affect the probability or severity of major defects in the pipeline network, and the target variables are used to indicate the probability or severity of major defects that may occur in the pipeline network within the first area to be identified.

[0202] The processing module 302 is further configured to train the pre-trained model based on the predicted variable and the target variable to obtain the pipeline defect prediction model.

[0203] Figure 4 A schematic diagram of the structure of the wastewater pipe network major defect identification equipment provided in this application. ​ As shown, this application provides a major defect identification device for sewage pipe networks. The major defect identification device 400 for sewage pipe networks includes: a receiver 401, a transmitter 402, a processor 403, and a memory 404.

[0204] Receiver 401 is used to receive instructions and data;

[0205] Transmitter 402 is used to send commands and data;

[0206] Memory 404 is used to store instructions executed by the computer;

[0207] The processor 403 is used to execute computer execution instructions stored in the memory 404 to implement the various steps of the sewage pipe network major defect identification method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the sewage pipe network major defect identification method.

[0208] Optionally, the memory 404 can be either standalone or integrated with the processor 403.

[0209] When the memory 404 is set up independently, the electronic device also includes a bus for connecting the memory 404 and the processor 403.

[0210] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the sewage pipe network major defect identification method performed by the aforementioned sewage pipe network major defect identification device.

[0211] This application also provides a program product, including a computer program, which, when executed by a processor, implements the method for identifying major defects in sewage pipe networks as described above.

[0212] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0213] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0214] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0215] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0216] When integrated units / modules are implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc.

[0217] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0218] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as these combinations of technical features do not contradict each other, they should be considered within the scope of this specification.

[0219] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0220] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying major defects in sewage pipe networks, characterized in that, The method includes: Acquire multiple first areas to be identified for the city to be identified, a multivariate dataset, and a pipeline defect prediction model. The multivariate dataset is used to indicate a set of multidimensional geographic information data of the city to be identified. The first areas to be identified include multiple second areas to be identified. The pipeline defect prediction model is controlled to perform initial defect prediction on the multiple first areas to be identified according to the multivariate dataset to obtain a first target area. The target area is used to indicate the area in the city to be identified where there are major pipeline defects. The pipeline defect prediction model is controlled to perform secondary defect prediction on multiple second unidentified areas in the first target area to obtain the second target area of ​​the unidentified city. Differential defect identification is performed on the second target area to obtain the target defect pipeline and the major defects of the target defect pipeline; Before obtaining the multivariate dataset of the city to be identified, the first area to be identified, and the pipeline defect prediction model, the method further includes: Acquire first geographic information and second geographic information, wherein the geographic information is used to perform multi-level division of the first region to be identified; The city to be identified is divided into regions according to the first geographic information to obtain the plurality of first regions to be identified. The plurality of first regions to be identified are divided into regions according to the second geographic information to obtain the plurality of second regions to be identified. The differential defect identification of the second target region to obtain the target defect pipeline and the major defects of the target defect pipeline includes: Obtain first water level information and second water level information of the second target area. The first water level information is used to indicate the groundwater level in the second target area, and the second water level information is used to determine the water level in the pipeline in the second target area. Based on the first water level information, a first identification mode for the differentiated defect identification is determined, and pipeline defect identification is performed on the second target area according to the first identification mode to obtain the target defect pipeline. The first identification mode is used to indicate the identification of defects outside the pipeline in the second target area. Based on the second water level information, a second identification mode for the differentiated defect identification is determined, and the target defect pipeline is identified according to the second identification mode to obtain the major defects of the target defect pipeline. The second identification mode is used to indicate the identification of defects within the pipeline in the second target area.

2. The method according to claim 1, characterized in that, The method of controlling the pipeline defect prediction model to perform initial defect prediction on the multiple first unidentified regions according to the multivariate dataset, to obtain the first target region, includes: First region identifiers of the plurality of first regions to be identified are obtained respectively, and the multivariate dataset is classified according to the plurality of first region identifiers to obtain a plurality of first region datasets. The region identifiers are used to distinguish the data information of the plurality of regions to be identified. The pipeline defect prediction model is controlled to perform defect prediction on the multiple first regional datasets respectively, and the first defect prediction value of the multiple first regions to be identified is obtained. The defect prediction value is used to indicate the proportion of major pipeline defects in different regions of the city to be identified. Obtain the defect prediction threshold of the city to be identified, and perform initial screening on multiple first defect prediction values ​​according to the defect prediction threshold to obtain the first target area. The defect prediction threshold is used to indicate the lower limit of the proportion of major pipeline defects in the area to be identified.

3. The method according to claim 2, characterized in that, The method of controlling the pipeline defect prediction model to perform secondary defect prediction on multiple second unidentified areas in the first target area to obtain the second target area of ​​the unidentified city includes: Obtain the second region identifiers of multiple second regions to be identified within the first target region; The first region dataset of the first target region is classified according to the second region identifier to obtain a second region dataset of multiple second regions to be identified in the first target region. The multiple second region datasets are input into the pipeline defect prediction model to obtain the second defect prediction values ​​of multiple second unidentified regions in the first target region; The multiple second defect prediction values ​​are further filtered according to the defect prediction threshold to obtain the second target area of ​​the city to be identified.

4. The method according to claim 1, characterized in that, The method further includes: Obtain pipeline information within the second target area and invoke the pipeline defect prediction model, which is used to predict major defects existing in the area where the target defect pipeline is located; The pipeline defect prediction model is controlled to predict pipeline defects based on the pipeline information. Based on the pipeline defect prediction results, the major defects of the target defect pipeline are corrected, and the correction results are re-determined as the major defects of the target defect pipeline.

5. The method according to claim 4, characterized in that, The method further includes: Obtain a pre-trained model and the historical multivariate dataset of the city to be identified, wherein the pre-trained model is used to describe the development trend present in the data arranged in chronological order; The historical multivariate dataset is analyzed and processed to obtain predictive variables and target variables for the city to be identified. The predictive variables are used to indicate the factors that affect the probability or severity of major defects in the pipeline network, and the target variables are used to indicate the probability or severity of major defects that may occur in the pipeline network in the first area to be identified. The pre-trained model is trained based on the predicted variables and the target variables to obtain the pipeline defect prediction model.

6. A device for identifying major defects in sewage pipe networks, characterized in that, include: The acquisition module is used to acquire multiple first areas to be identified, a multivariate dataset, and a pipeline defect prediction model for a city to be identified. The multivariate dataset is used to indicate a collection of multidimensional geographic information data of the city to be identified. The first areas to be identified include multiple second areas to be identified. The processing module is used to control the pipeline defect prediction model to perform initial defect prediction on the multiple first areas to be identified according to the multivariate dataset, and obtain a first target area. The target area is used to indicate the area in the city to be identified where there are major pipeline defects. The processing module is also used to control the pipeline defect prediction model to perform secondary defect prediction on multiple second areas to be identified in the first target area, so as to obtain the second target area of ​​the city to be identified. The processing module is also used to perform differentiated defect identification on the second target area to obtain the target defect pipeline and the major defects of the target defect pipeline; The acquisition module is further configured to acquire first geographic information and second geographic information, wherein the geographic information is used to perform multi-level division of the first area to be identified. The processing module is further configured to divide the city to be identified into regions according to the first geographic information to obtain the plurality of first regions to be identified; The processing module is further configured to divide the plurality of first regions to be identified into regions according to the second geographic information, thereby obtaining the plurality of second regions to be identified; The acquisition module is further configured to acquire first water level information and second water level information of the second target area, wherein the first water level information is used to indicate the groundwater level in the second target area and the second water level information is used to determine the water level in the pipeline in the second target area. The device further includes: a determining module; The determining module is used to determine the first identification mode of the differentiated defect identification based on the first water level information. The processing module is further configured to identify pipeline defects in the second target area according to the first identification mode to obtain the target defect pipeline, wherein the first identification mode is used to indicate the identification of defects outside the pipeline in the second target area. The determining module is further configured to determine the second identification mode for the differentiated defect identification based on the second water level information; The processing module is further configured to identify pipeline defects in the target defective pipeline according to the second identification mode, thereby obtaining major defects in the target defective pipeline. The second identification mode is used to indicate pipeline defect identification in the second target area.

7. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 5.

Images