Pipeline internal sewage detection system for communication

Abstract

The invention relates to the technical field of sewage detection, and particularly discloses a pipeline internal sewage detection system for communication, which comprises an information acquisition module for acquiring sewage data of each preset point location of a pipeline and image information of each preset point location of the pipeline; the data processing module performs target feature screening on the sewage data of each preset point location to obtain abnormal sewage features, determines the acquisition point location of the abnormal sewage features as an abnormal point location and determines image information of the abnormal point location; carrying out risk analysis on the image information of each abnormal point location and the corresponding abnormal sewage features, and obtaining pipeline flow velocity parameters of the abnormal point locations; the analysis module performs pipeline sewage distribution analysis based on the abnormal sewage characteristics, the image information of the abnormal point locations and the pipeline flow velocity parameters of the abnormal point locations, and judges whether current pipeline problem information matching is correct or not according to a pipeline sewage distribution analysis result: if yes, the pipeline problem information is output; and if not, monitoring the current abnormal point location data in real time.

Description

Technical Field

[0001] The invention relates to the technical field of sewage detection, and in particular to a sewage detection system inside a communication pipeline. Background Art

[0002] With the development of communication technology, the number and complexity of communication pipelines are increasing, and the demand for monitoring the sewage environment inside pipelines is also growing. Traditional manual detection methods are inefficient and have safety risks. Therefore, automated and intelligent sewage detection systems have become a research hotspot.

[0003] The internal sewage detection system for communication pipes is widely used in urban drainage systems, industrial wastewater treatment, sewage treatment plants and other fields; the intelligent sewage detection system can monitor the water level, water quality, flow rate and other parameters inside the pipe in real time, and promptly detect and deal with potential sewage leakage problems.

[0004] Currently, sewage detection inside communication pipes usually only involves the detection of a single situation of sewage or pipeline problems, and rarely conducts a detailed analysis based on the correlation between sewage changes and pipeline problems. Therefore, the impact of chemical, physical and biological factors of sewage quality on the inner wall of the pipe and thus the flow rate in the pipe is often ignored, resulting in early warning communication and subsequent maintenance work to deal with pipeline sewage problems after pipeline problems occur, resulting in unnecessary economic losses and manpower costs. Summary of the invention

[0005] The purpose of the present invention is to provide a sewage detection system for communication pipelines to solve the following technical problems:

[0006] How to achieve timely distribution of information on pipeline problems caused by sewage in the pipeline and pipeline safety early warning process through signal communication and analysis of sewage and pipeline problems.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A sewage detection system for communication pipelines, comprising:

[0009] The information acquisition module is used to obtain the sewage data of each preset point in the pipeline and the image information of each preset point in the pipeline; the sewage data includes sewage content parameters, sewage water quality parameters and sewage temperature parameters; the image information is used to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline and the internal environment information of the pipeline;

[0010] A data processing module is used to perform target feature screening on the sewage data of each preset point to obtain abnormal sewage features and to identify the collection points of abnormal sewage features as abnormal points and determine the image information of the abnormal points;

[0011] and for conducting risk analysis on the image information of each abnormal point and the corresponding abnormal sewage characteristics, and obtaining the pipeline flow velocity parameters of the abnormal point;

[0012] The analysis module is used to perform pipeline sewage distribution analysis based on abnormal sewage characteristics, image information of abnormal points, and pipeline flow velocity parameters of abnormal points, and to determine whether the current pipeline problem information match is correct according to the pipeline sewage distribution analysis results:

[0013] If yes, then output the pipeline problem information;

[0014] If not, the current abnormal point data is monitored in real time;

[0015] The monitoring module is used to upload pipeline sewage change trend information to the control center through real-time monitoring results, and generate early warning signals based on the sewage change trend information.

[0016] Preferably, the data processing module includes:

[0017] Collect the sewage content parameters, sewage water quality parameters and sewage temperature parameters of each preset point for data cleaning and pre-processing of outlier removal;

[0018] The pre-processed sewage content parameters, sewage water quality parameters, sewage temperature parameters and image information of each preset point are input into the convolutional neural network model for learning and training to obtain the feature ranking of the importance of each parameter;

[0019] The features with the top n importance of each parameter are extracted as the target features, and the parameters corresponding to the target features are compared with the preset standard thresholds:

[0020] If it is greater than the preset standard threshold, it is determined to be an abnormal sewage feature and the preset point where it is located is marked as an abnormal point.

[0021] Preferably, the data processing module further includes:

[0022] The image information of each abnormal point and the corresponding abnormal sewage characteristics are input into the random forest model for risk assessment to obtain the risk coefficient of the abnormal sewage characteristics;

[0023] The risk factors of abnormal sewage characteristics include the risk factor of abnormal sewage content, the risk factor of abnormal sewage quality, and the risk factor of abnormal sewage temperature;

[0024] Compare the risk factor of abnormal sewage characteristics with the preset risk factor threshold:

[0025] If the risk factor of the abnormal sewage characteristics is greater than the preset risk factor threshold, the pipeline flow velocity parameters of the abnormal point are obtained.

[0026] Preferably, the analysis module comprises:

[0027] By formula Calculate the distribution coefficient Par of the i-th abnormal point i ;

[0028] Among them, f is the preset allocation function; Ev1 is the risk coefficient of abnormal sewage content; Ev2 is the risk coefficient of abnormal sewage quality; Ev3 is the risk coefficient of abnormal sewage temperature; ha i is the abnormal sewage content parameter of the ith abnormal point; sz i Abnormal sewage water quality parameters of the ith abnormal point; wd i is the abnormal sewage temperature parameter of the ith abnormal point; fp i is the pipeline velocity parameter of the i-th abnormal point; ε i is the location coefficient of the ith abnormal point.

[0029] Preferably, the allocation coefficient Par of the current i-th abnormal point is i and the preset threshold interval [Par A ,Par B ] for comparison:

[0030] If Par i ∈[Par A ,Par B ], it is judged that the pipeline at the current abnormal point is in a safe range;

[0031] like It is judged that the pipeline at the current abnormal point is in the risk range and there is a pipeline problem information risk.

[0032] Preferably, the analysis module further includes collecting real-time image information of abnormal points where there is a risk of pipeline problem information;

[0033] Identify the real-time image information based on the pipeline problem information in the historical database to determine whether the current pipeline problem information matches correctly:

[0034] If yes, then output the pipeline problem information;

[0035] If not, monitor the current abnormal point in real time;

[0036] Pipeline problem information includes pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

[0037] Preferably, the monitoring module comprises:

[0038] Obtain the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters of each abnormal point within a preset time period;

[0039] Compare the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters with the standard time variation curves of each parameter in the same coordinate system to determine the size of the overlap area of ​​each curve:

[0040] If any of the overlapping areas is smaller than the corresponding preset overlapping area threshold, it is determined that there is a pipeline problem and a pipeline warning message is generated;

[0041] If all overlapping areas are larger than the corresponding preset overlapping area threshold, it is determined that there is no pipeline problem and abnormal point monitoring continues.

[0042] Preferably, it also includes:

[0043] The probability of pipeline problem information is estimated based on the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline displayed by the image information of the abnormal points in the pipeline.

[0044] Preferably, the pipeline problem information probability estimation method is:

[0045] Collect image samples of abnormal points in the pipeline to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline, obtain the text information of the image samples for large database scene matching, and obtain the keywords of the scene problem;

[0046] Determine the overlap probability of the scenario problem keywords selected by the big database as the pipeline problem information parameter;

[0047] Statistical pipeline problem information parameter table, filter from the table to obtain the probability of pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

[0048] Beneficial effects of the present invention:

[0049] (1) The present invention implements preprocessing and screening of target features of sewage data acquired by the information acquisition module by setting a data processing module, determines and analyzes abnormal information by screening the target features of sewage data, and determines image information of abnormal points by identifying position signals of abnormal sewage features for the abnormal information of sewage data, and performs risk analysis on the image information of each abnormal point and the corresponding abnormal sewage features, determines the influence of the abnormal sewage features on the pipeline flow rate by judging the analysis results, obtains the pipeline flow rate parameters of the abnormal points, and then performs further pipeline safety prediction based on the current pipeline flow rate parameters.

[0050] (2) The present invention realizes matching analysis of abnormal sewage characteristics, image information of abnormal points and pipeline flow velocity parameters of abnormal points by setting up an analysis module, and judges whether there is an actual problem in the current pipeline based on the analysis results, and judges whether the current pipeline problem information is matched correctly. If the judgment result is yes, the pipeline problem information is output; if not, the current abnormal point data is monitored in real time. The matching analysis process realizes further prediction of the impact on the sewage inside the pipeline and timely identification of pipeline problem information risks.

[0051] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0053] Figure 1 This is a module diagram of a sewage detection system inside a communication pipeline according to the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] See also Figure 1 As shown, the present invention is a sewage detection system for communication pipelines, comprising:

[0056] The information acquisition module is used to obtain the sewage data of each preset point in the pipeline and the image information of each preset point in the pipeline; the sewage data includes sewage content parameters, sewage water quality parameters and sewage temperature parameters; the image information is used to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline and the internal environment information of the pipeline;

[0057] A data processing module is used to perform target feature screening on the sewage data of each preset point to obtain abnormal sewage features and to identify the collection points of abnormal sewage features as abnormal points and determine the image information of the abnormal points;

[0058] and for conducting risk analysis on the image information of each abnormal point and the corresponding abnormal sewage characteristics, and obtaining the pipeline flow velocity parameters of the abnormal point;

[0059] The analysis module is used to perform pipeline sewage distribution analysis based on abnormal sewage characteristics, image information of abnormal points, and pipeline flow velocity parameters of abnormal points, and to determine whether the current pipeline problem information match is correct according to the pipeline sewage distribution analysis results:

[0060] If yes, then output the pipeline problem information;

[0061] If not, the current abnormal point data is monitored in real time;

[0062] The monitoring module is used to upload pipeline sewage change trend information to the control center through real-time monitoring results, and generate early warning signals based on the sewage change trend information.

[0063] In the above technical solution, the present invention realizes the systematization process of sewage detection inside the communication pipeline by setting up an information collection module, a data processing module, an analysis module and a monitoring module. Specifically, firstly, various types of information inside the pipeline are collected by the information collection module. For the equipment for collection, this embodiment uses a common combination sensor, which is often used to receive sewage data and image information of each preset point, and obtain sewage data of the pipeline, wherein the sewage data includes sewage content parameters, sewage water quality parameters and sewage temperature parameters; the image information inside the pipeline is used to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline and the internal environment information of the pipeline.

[0064] Then, a data processing module is set up to preprocess the sewage data obtained by the information collection module and screen the target features. The abnormal information is determined and analyzed by screening the target features of the sewage data. The abnormal information of the sewage data includes abnormal sewage characteristics. The image information of the abnormal points is determined by identifying the position signals of the abnormal sewage characteristics. Risk analysis is performed on the image information of each abnormal point and the corresponding abnormal sewage characteristics. The analysis results are judged to determine the impact of the abnormal sewage characteristics on the pipeline flow velocity, and the pipeline flow velocity parameters of the abnormal points are obtained, and then further pipeline safety prediction is performed based on the current pipeline flow velocity parameters.

[0065] Next, an analysis module is set up to match and analyze the abnormal sewage characteristics, image information of abnormal points, and pipeline flow velocity parameters of abnormal points, and judge whether there is an actual problem in the current pipeline based on the analysis results. Specifically, it is judged whether the current pipeline problem information match is correct. If so, the pipeline problem information is output; if not, the current abnormal point data is monitored in real time.

[0066] Finally, by setting up the monitoring module to obtain the analysis module real-time monitoring results, upload and analyze the changes in piped sewage, analyze the changing trends of piped sewage, and generate early warning signals for situations that exceed normal piped sewage.

[0067] As an embodiment of the present invention, the data processing module includes:

[0068] Collect the sewage content parameters, sewage water quality parameters and sewage temperature parameters of each preset point for data cleaning and pre-processing of outlier removal;

[0069] The pre-processed sewage content parameters, sewage water quality parameters, sewage temperature parameters and image information of each preset point are input into the convolutional neural network model for learning and training to obtain the feature ranking of the importance of each parameter;

[0070] The features with the top n importance of each parameter are extracted as the target features, and the parameters corresponding to the target features are compared with the preset standard thresholds:

[0071] If it is greater than the preset standard threshold, it is determined to be an abnormal sewage feature and the preset point where it is located is marked as an abnormal point.

[0072] In the above technical scheme, the method of data processing in this embodiment specifically includes first preprocessing the sewage content parameters, sewage water quality parameters and sewage temperature parameters of the designed sewage information of the preset points, and inputting the preprocessed sewage data and the image information related to the sewage into the convolutional neural network model for machine learning, training multiple times to obtain the feature sequence of each parameter and obtain feature sorting information; screening the important factors that cause changes in sewage inside the pipeline in historical situations, selecting the size of n according to the actual influencing conditions of the pipeline in actual operation, and then extracting the features of the importance of the top n in the sorting as the target features, and analyzing the parameter sizes corresponding to the target features, marking the situation that exceeds the standard threshold as an abnormal sewage feature, and considering the marked preset point as an abnormal point.

[0073] As an embodiment of the present invention, the data processing module further includes:

[0074] The image information of each abnormal point and the corresponding abnormal sewage characteristics are input into the random forest model for risk assessment to obtain the risk coefficient of the abnormal sewage characteristics;

[0075] The risk factors of abnormal sewage characteristics include the risk factor of abnormal sewage content, the risk factor of abnormal sewage quality, and the risk factor of abnormal sewage temperature;

[0076] Compare the risk factor of abnormal sewage characteristics with the preset risk factor threshold:

[0077] If the risk factor of the abnormal sewage characteristics is greater than the preset risk factor threshold, the pipeline flow velocity parameters of the abnormal point are obtained.

[0078] In the above technical solution, since there are situations that may affect pipeline safety and situations that have little or no impact on pipeline safety in the marked abnormal points and abnormal sewage feature information; this requires constructing the risk of the current abnormal sewage feature and constructing the risk coefficient based on the pipeline feature. In this embodiment, a random forest model is required for risk assessment, and the abnormal data including the image information of the abnormal points and the abnormal sewage features are evaluated for risk assessment, thereby obtaining the risk coefficient of the abnormal sewage features. Different features of the risk coefficient of the abnormal sewage features are divided into abnormal sewage content risk coefficient, abnormal sewage water quality risk coefficient, and abnormal sewage temperature risk coefficient.

[0079] Moreover, the image information of the abnormal points here includes the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline that may or may not affect the safety of the pipeline. For example, obvious cracks, depressions, or abnormal protrusions appear at the internal structure of the pipeline and the sewage distribution position, which may affect the safety of the pipeline, and the sewage contact at the sewage distribution position inside the pipeline affects the color of the inner wall of the pipeline, which may not affect the safety of the pipeline; in addition, it also includes the abnormal increase or decrease of the internal environment temperature of the pipeline, which may affect the safety of the pipeline. The above situations only describe some situations in the actual image and are not limited to other actual situations described above.

[0080] The model in this design conducts evaluation and analysis to obtain the risk coefficient of abnormal sewage characteristics, and judges the risk coefficient by comparing the risk coefficient of abnormal sewage characteristics with the preset risk coefficient threshold. If the risk coefficient of abnormal sewage characteristics is greater than the preset risk coefficient threshold, the pipeline flow velocity parameters of the abnormal point are obtained, and the pipeline hazard is predicted in time according to the pipeline flow velocity.

[0081] Among them, pipeline flow rate parameters refer to a series of physical quantities used to describe the fluid movement state during pipeline transportation; they mainly include flow rate, flow velocity, pressure, etc. The pipeline flow rate parameters are obtained by fitting the flow rate, flow velocity, and pressure physical quantities of the pipeline through a combination of sensors at preset points.

[0082] As an embodiment of the present invention, the analysis module includes:

[0083] By formula Calculate the distribution coefficient Par of the i-th abnormal point i ;

[0084] Among them, f is the preset allocation function; Ev1 is the risk coefficient of abnormal sewage content; Ev2 is the risk coefficient of abnormal sewage quality; Ev3 is the risk coefficient of abnormal sewage temperature; ha i is the abnormal sewage content parameter of the ith abnormal point; sz i Abnormal sewage water quality parameters of the ith abnormal point; wd i is the abnormal sewage temperature parameter of the ith abnormal point; fp i is the pipeline velocity parameter of the i-th abnormal point; ε i is the location coefficient of the ith abnormal point.

[0085] In the above technical solution, in this embodiment, the pipeline flow rate parameters obtained by data processing are further predicted by the analysis module, specifically by calculating the formula Calculate the distribution coefficient Par of the i-th abnormal point i , according to the distribution coefficient Par of different abnormal points i The actual risk of the pipeline at the abnormal point can be judged, that is, the comprehensive size of the abnormal sewage content risk, abnormal sewage water quality risk, and abnormal sewage temperature risk and the pipeline flow rate ratio reflects the change of unit content in the pipeline, and the distribution of the abnormal point is determined. i The impact of size on sewage inside the pipeline was further predicted, and the risk of pipeline problem information was identified in a timely manner.

[0086] It should be explained that the preset allocation function f is a task allocation function pre-set according to the historical sewage data, which is obtained according to the allocation optimization algorithm, thereby ensuring that the result of the allocation coefficient is within a reasonable range; the position coefficient ε of the i-th abnormal point i The settings are selected based on the importance of different abnormal points on the pipeline and will not be described in detail here.

[0087] As an implementation mode of the present invention, the allocation coefficient Par of the current i-th abnormal point is i and the preset threshold interval [Par A ,Par B ] for comparison:

[0088] If Par i ∈[Par A ,Par B ], it is judged that the pipeline at the current abnormal point is in a safe range;

[0089] like It is judged that the pipeline at the current abnormal point is in the risk range and there is a pipeline problem information risk.

[0090] In the above technical solution, according to the analysis of the current distribution coefficient Par of the i-th abnormal point i Whether the size is within the preset standard range, the danger level of the pipeline at the current abnormal point is judged, and then the risk tracking and data monitoring process of pipeline problem information is carried out.

[0091] As an embodiment of the present invention, the analysis module further includes collecting real-time image information of abnormal points where there is a risk of pipeline problem information;

[0092] Identify the real-time image information based on the pipeline problem information in the historical database to determine whether the current pipeline problem information matches correctly:

[0093] If yes, then output the pipeline problem information;

[0094] If not, monitor the current abnormal point in real time;

[0095] Pipeline problem information includes pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

[0096] In the above technical solution, in this embodiment, the real-time image information of the acquired abnormal points is further collected to realize the timely tracking and early warning process of risks. Specifically, the real-time image information is identified according to the pipeline problem information in the historical database to determine whether the current pipeline problem information matches correctly. If it is correct, the pipeline problem information is output; wherein the pipeline problem information generally includes pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage, otherwise the current abnormal point is monitored.

[0097] As an embodiment of the present invention, the monitoring module includes:

[0098] Obtain the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters of each abnormal point within a preset time period;

[0099] Compare the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters with the standard time variation curves of each parameter in the same coordinate system to determine the size of the overlap area of ​​each curve:

[0100] If any of the overlapping areas is smaller than the corresponding preset overlapping area threshold, it is determined that there is a pipeline problem and a pipeline warning message is generated;

[0101] If all overlapping areas are larger than the corresponding preset overlapping area threshold, it is determined that there is no pipeline problem and abnormal point monitoring continues.

[0102] In the above technical scheme, this embodiment dynamically monitors the time change curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters, and respectively determines their similarities with the standard time change curves under the normal pipeline detection state, and analyzes the size of the overlapping area of ​​the curves. The larger the overlapping area, the more it proves that their changing trends are the same, and then it can be judged as a normal operation condition based on the current changing trend. Otherwise, it is judged that there is a pipeline problem, and early warning information of the pipeline is generated at the corresponding position.

[0103] As an embodiment of the present invention, it also includes:

[0104] The probability of pipeline problem information is estimated based on the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline displayed by the image information of the abnormal points in the pipeline.

[0105] In the above technical solution, the probability estimation of pipeline problem information helps to further analyze the severity of the pipeline problem, which in turn helps to accurately predict the occurrence of the pipeline problem.

[0106] As an implementation mode of the present invention, the pipeline problem information probability estimation method is:

[0107] Collect image samples of abnormal points in the pipeline to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline, obtain the text information of the image samples for large database scene matching, and obtain the keywords of the scene problem;

[0108] Determine the overlap probability of the scenario problem keywords selected by the big database as the pipeline problem information parameter;

[0109] Statistical pipeline problem information parameter table, filter from the table to obtain the probability of pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

[0110] In the above technical scheme, first, image samples of abnormal points in the pipeline are collected by professional equipment, and the image samples include the internal structure of the pipeline, the distribution of sewage inside the pipeline, and the internal environmental information of the pipeline; then, the collected image samples are processed to extract relevant text information for further analysis; then, the extracted text information is matched with the scene problem keywords in the big database to identify and classify pipeline problems; further, the overlap probability between the selected scene problem keywords is calculated, and this probability can be used as a measure of the pipeline problem information parameters; and by counting and sorting the pipeline problem information parameters, a probability statistical table is created to help screen and identify specific pipeline problems, such as blockage, corrosion, deformation and leakage. Finally, based on the data in the probability statistical table, the probability of occurrence of various pipeline problems is calculated.

[0111] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0112] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this application, they shall all fall within the protection scope of the present invention.

Claims

1. A sewage detection system for communication pipelines, characterized in that: include: An information collection module is used to obtain sewage data and image information of each preset point in the pipeline; the sewage data includes sewage content parameters, sewage water quality parameters and sewage temperature parameters; the image information is used to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline and the internal environment information of the pipeline; A data processing module is used to perform target feature screening on the sewage data of each preset point to obtain abnormal sewage features and to identify the collection points of abnormal sewage features as abnormal points and determine the image information of the abnormal points; and for conducting risk analysis on the image information of each abnormal point and the corresponding abnormal sewage characteristics, and obtaining the pipeline flow velocity parameters of the abnormal point; The analysis module is used to perform pipeline sewage distribution analysis based on abnormal sewage characteristics, image information of abnormal points, and pipeline flow velocity parameters of abnormal points, and to determine whether the current pipeline problem information match is correct according to the pipeline sewage distribution analysis results: If yes, then output the pipeline problem information; If not, the current abnormal point data is monitored in real time; The monitoring module is used to upload pipeline sewage change trend information to the control center through real-time monitoring results, and generate early warning signals based on the sewage change trend information.

2. A communication pipeline internal sewage detection system according to claim 1, characterized in that: The data processing module comprises: Collect the sewage content parameters, sewage water quality parameters and sewage temperature parameters of each preset point for data cleaning and pre-processing of outlier removal; The pre-processed sewage content parameters, sewage water quality parameters, sewage temperature parameters and image information of each preset point are input into the convolutional neural network model for learning and training to obtain the feature ranking of the importance of each parameter; The features with the top n importance of each parameter are extracted as the target features, and the parameters corresponding to the target features are compared with the preset standard thresholds: If it is greater than the preset standard threshold, it is determined to be an abnormal sewage feature and the preset point where it is located is marked as an abnormal point.

3. A communication pipeline internal sewage detection system according to claim 2, characterized in that: The data processing module also includes: The image information of each abnormal point and the corresponding abnormal sewage characteristics are input into the random forest model for risk assessment to obtain the risk coefficient of the abnormal sewage characteristics; The risk coefficients of abnormal sewage characteristics include abnormal sewage content risk coefficient, abnormal sewage water quality risk coefficient, and abnormal sewage temperature risk coefficient; Compare the risk factor of abnormal sewage characteristics with the preset risk factor threshold: If the risk factor of the abnormal sewage characteristics is greater than the preset risk factor threshold, the pipeline flow velocity parameters of the abnormal point are obtained.

4. A communication pipeline internal sewage detection system according to claim 3, characterized in that: The analysis module comprises: By formula Calculate the distribution coefficient Par of the i-th abnormal point i ; Among them, f is the preset allocation function; Ev1 is the risk coefficient of abnormal sewage content; Ev2 is the risk coefficient of abnormal sewage quality; Ev3 is the risk coefficient of abnormal sewage temperature; ha i is the abnormal sewage content parameter of the ith abnormal point; sz i Abnormal sewage water quality parameters of the ith abnormal point; wd i is the abnormal sewage temperature parameter of the ith abnormal point; fp i is the pipeline velocity parameter of the i-th abnormal point; ε i is the location coefficient of the ith abnormal point.

5. A communication pipeline internal sewage detection system according to claim 4, characterized in that: The distribution coefficient Par of the current i-th abnormal point i and the preset threshold interval [Par A ,Par B ] for comparison: If Par i ∈[Par A ,Par B ], it is judged that the pipeline at the current abnormal point is in a safe range; like It is judged that the pipeline at the current abnormal point is in the risk range and there is a pipeline problem information risk.

6. A communication pipeline internal sewage detection system according to claim 5, characterized in that: The analysis module also includes collecting real-time image information of abnormal points with pipeline problem information risks; Identify the real-time image information based on the pipeline problem information in the historical database to determine whether the current pipeline problem information matches correctly: If yes, then output the pipeline problem information; If not, monitor the current abnormal point in real time; The pipeline problem information includes pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

7. A communication pipeline internal sewage detection system according to claim 1, characterized in that: The monitoring module comprises: Obtain the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters of each abnormal point within a preset time period; Compare the time variation curves of abnormal sewage content parameters, abnormal sewage water quality parameters, and abnormal sewage temperature parameters with the standard time variation curves of each parameter in the same coordinate system to determine the size of the overlap area of ​​each curve: If any of the overlapping areas is smaller than the corresponding preset overlapping area threshold, it is determined that there is a pipeline problem and a pipeline warning message is generated; If all overlapping areas are larger than the corresponding preset overlapping area threshold, it is determined that there is no pipeline problem and abnormal point monitoring continues.

8. A communication pipeline internal sewage detection system according to claim 6, characterized in that: Also includes: The probability of pipeline problem information is estimated based on the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline displayed by the image information of the abnormal points in the pipeline.

9. A communication pipeline internal sewage detection system according to claim 8, characterized in that: The pipeline problem information probability estimation method is: Collect image samples of abnormal points in the pipeline to display the internal structure information of the pipeline, the sewage distribution information inside the pipeline, and the internal environment information of the pipeline, obtain the text information of the image samples for large database scene matching, and obtain the keywords of the scene problem; Determine the overlap probability of the scenario problem keywords selected by the big database as the pipeline problem information parameter; Statistical pipeline problem information parameter table, filter from the table to obtain the probability of pipeline blockage, pipeline corrosion, pipeline deformation, and pipeline leakage.

Images