Urban drainage pipe network monitoring and management system based on Internet of Things

Abstract

The invention relates to an urban drainage pipe network monitoring and management system based on the Internet of Things. The system comprises a data acquisition module which acquires underground environment data acquired by a pipe network sensor to form an initial environment data set; the configuration optimization module classifies the data set through a support vector machine algorithm to obtain an environment type label, further optimizes the mounting position and the fixing mode of the sensor, and determines an adaptive sensor configuration scheme; a transmission optimization module predicts a transmission path and screens a low-attenuation path through a random forest algorithm based on the scheme to obtain a stable transmission data stream; and the exception handling module extracts data stream feature vectors, potential exceptions are analyzed and marked through an exception detection model, a list is formed, adaptive control response signals are generated after exception levels are divided, and a final monitoring report is output in combination with action feedback data and a processing track. According to the system, the data acquisition accuracy and the sensor adaptability are improved, and pipe network abnormity rapid identification, accurate classification and efficient disposal are realized.

Description

Technical Field

[0001] This invention belongs to the field of infrastructure monitoring technology, and in particular relates to an Internet of Things-based urban drainage network monitoring and management system. Background Technology

[0002] As a core lifeline project ensuring the normal operation of a city, the stability of urban drainage networks directly affects the safety of municipal infrastructure, the quality of life of residents, and the sustainability of the ecological environment. Currently, traditional drainage network monitoring relies heavily on manual inspections and fixed-point sampling, which suffers from limited coverage, delayed data collection, and high labor costs. It also struggles to capture potential risks such as changes in underground soil pressure and leakage in the network in real time. Even with the introduction of IoT sensing technology in some solutions, technical bottlenecks remain, including a lack of dynamic adaptability of sensor configurations to different underground environments, signal attenuation and interference leading to data distortion in complex soil media, and reliance on single static thresholds for anomaly detection without corresponding analysis of the network's operational status. These limitations fail to meet the urgent needs of modern cities for refined and intelligent operation and maintenance management of drainage networks. Summary of the Invention

[0003] Therefore, it is necessary to provide an IoT-based urban drainage network monitoring and management system that can improve the accuracy of data acquisition and the adaptability of sensor installation, and enable rapid identification, accurate classification and efficient handling of network anomalies, in response to the above-mentioned technical problems.

[0004] Firstly, this application provides an Internet of Things-based urban drainage network monitoring and management system, including:

[0005] The data acquisition module is used to acquire underground environmental data collected by sensors deployed in the urban drainage network to obtain an initial environmental dataset.

[0006] The configuration optimization module is used to classify underground environment types based on the initial environment dataset using the support vector machine algorithm to obtain environment type labels; it is also used to adjust the sensor installation position and fixing method based on the environment type labels to determine the sensor configuration scheme that is suitable for the current burial depth.

[0007] The transmission optimization module is used to predict the transmission path based on the sensor configuration scheme using the random forest algorithm to obtain an optimized transmission path set. If there is a low-attenuation path in the optimized transmission path set, the low-attenuation path is selected for signal transmission to obtain a stable transmission data stream.

[0008] The anomaly handling module is used to analyze the feature vectors in the stable transmission data stream using an anomaly detection model. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined. It is also used to classify the anomaly level based on the list of anomalies and generate control response signals adapted to the pipeline network operation status. Furthermore, it is used to obtain the final monitoring report based on the action feedback data after the response signals are executed and the anomaly event handling trajectory.

[0009] In one embodiment, the configuration optimization module is further configured to:

[0010] Soil pressure data was extracted from the initial environmental dataset. Based on the soil pressure data, the support vector machine algorithm was used to classify the underground environment types, and the underground environment classification results were obtained.

[0011] Based on the underground environment classification results and the preset soil pressure threshold, the soil pressure data is judged. If the soil pressure data exceeds the preset threshold, the corresponding underground environment is classified as a high-pressure environment, and an environment type label with environmental characteristics is generated.

[0012] Based on the underground environment characteristics corresponding to the environment type label, the initial installation parameters of the sensor are adjusted according to the geometric coverage theory and the goal of minimizing the monitoring blind zone, and optimized sensor installation location data is generated.

[0013] Based on the optimized sensor installation location data, combined with the current burial depth of the sensor, soil medium characteristics, and pressure tolerance requirements, the appropriate sensor fixing method parameters are calculated through a multi-objective optimization function.

[0014] The optimized sensor installation location data is coupled and integrated with the sensor fixing method parameters to obtain a sensor configuration scheme adapted to the underground geological environment.

[0015] In one embodiment, the multi-objective optimization function is expressed using the following formula:

[0016]

[0017] in, Represents the vector of optimization variables. , Indicates the anchoring depth. Indicates the stiffness of the fastener. Indicates the thickness of the protective layer. Indicates a fixed angle. , , Represents dynamic weighting coefficients based on environment type labels. Dynamic allocation Describe the objective function. , Indicates the upper limit of allowable displacement. , Indicates soil pressure, Indicates the contact area of ​​the fastener. Indicates the internal friction angle of the soil. This represents soil cohesion, set based on pressure tolerance requirements. Describe the objective function. , This indicates the maximum allowable adaptation deviation. , Indicates the current burial depth of the sensor. Indicates the dynamic adaptation coefficient of anchoring depth. Indicates the equivalent stiffness of the soil. , This indicates the optimal thickness of the protective layer. / 8, Indicates the wavelength of the sensor signal. Describe the objective function. , Indicates the maximum allowable attenuation of the signal. , This indicates the attenuation coefficient of the protective layer material, calibrated in conjunction with the characteristics of the soil medium. , This represents dynamic constraints. , , This represents the combined stress on the sensor's mounting structure under soil pressure. , This represents the normal stress borne by a fixed structure. , This indicates the cross-sectional area of ​​the fixed component under stress. This represents the shear stress borne by a fixed structure. , This indicates the allowable stress of the fastener material.

[0018] In one embodiment, the transmission optimization module is further configured to:

[0019] Obtain real-time environmental data acquisition results under the sensor configuration scheme.

[0020] The signal strength, soil medium attenuation coefficient, and path occlusion feature parameters from the collected results are input into the random forest algorithm to predict the appropriate signal transmission path, thus obtaining an optimized transmission path set.

[0021] The scattering parameter method is used to quantitatively analyze the dissipation attenuation and reflection attenuation of each path in the optimized transmission path set, and to determine whether there are low attenuation paths with attenuation below the preset signal attenuation threshold.

[0022] If a low-attenuation path exists in the optimized transmission path set, the low-attenuation path with the smallest attenuation is selected first for signal transmission. By reducing signal reflection and crosstalk through terminal impedance matching, a stable transmission data stream is obtained.

[0023] In one embodiment, the exception handling module further includes:

[0024] Anomaly detection unit, used for:

[0025] Static parameters including soil pressure and humidity, as well as feature vectors of signal transmission jitter rate and attenuation fluctuation rate, are extracted from the stable data stream.

[0026] An anomaly detection model trained on historical normal data streams is used to analyze feature vectors. If a feature vector exceeds the normal threshold range or the deviation rate exceeds a preset threshold, it is marked as a potential anomaly.

[0027] Record the occurrence time, pipeline topology location, parameter deviation magnitude, and associated monitoring nodes of potential anomalies to determine the list of abnormal events.

[0028] The exception handling unit is used for:

[0029] Based on the parameter deviation amplitude of each anomaly in the abnormal event list, the weight of key areas of the pipeline network, and the range of impact prediction, the analytic hierarchy process (AHP) is used to classify the anomaly levels.

[0030] By performing correlation analysis on the anomaly level, topology location, parameter type, real-time operating parameters of the pipeline network, and historical similar anomaly handling case library of abnormal events, differentiated control response signals including execution priority, target execution unit, and parameter adjustment threshold are generated.

[0031] The system acquires action feedback data after the control response signal is issued and executed, combines it with the handling trajectory of abnormal events, integrates the root cause analysis of abnormalities, differentiated response measures and the effect of improving operating conditions, and obtains the final monitoring report.

[0032] Secondly, this application also provides a method for monitoring and managing urban drainage networks based on the Internet of Things, the method including:

[0033] Obtain underground environmental data collected by sensors deployed in the urban drainage network to obtain an initial environmental dataset.

[0034] Based on the initial environmental dataset, the support vector machine algorithm is used to classify underground environmental types and obtain environmental type labels.

[0035] Adjust the sensor installation location and fixing method based on the environment type label to determine the sensor configuration scheme suitable for the current burial depth.

[0036] Based on the sensor configuration scheme, the random forest algorithm is used to predict the transmission path, resulting in an optimized set of transmission paths.

[0037] If a low-attenuation path exists in the optimized transmission path set, then the low-attenuation path is selected for signal transmission to obtain a stable data stream.

[0038] An anomaly detection model is used to analyze the feature vectors in a stable data stream. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined.

[0039] Based on the list of abnormal events, the abnormality levels are classified to generate control response signals that are adapted to the operating status of the pipeline network.

[0040] The final monitoring report is obtained based on the action feedback data after the response signal is executed and the abnormal event handling trajectory.

[0041] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned system.

[0042] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned system.

[0043] The aforementioned IoT-based urban drainage network monitoring and management system, computer equipment, and storage media include a data acquisition module that acquires underground environmental data collected by sensors deployed in the urban drainage network to form an initial environmental dataset. A configuration optimization module uses this initial environmental dataset as input, employs a support vector machine algorithm to classify underground environmental types, outputs environmental type labels, and then adjusts the sensor installation positions and fixing methods based on these labels to determine a sensor configuration scheme suitable for the current burial depth. A transmission optimization module, based on the sensor configuration scheme, predicts signal transmission paths using a random forest algorithm to generate an optimized transmission path set. If a low-attenuation path exists in the optimized transmission path set, it selects that path for signal transmission, resulting in a stable data stream. An anomaly handling module extracts feature vectors from the stable data stream, analyzes these feature vectors using an anomaly detection model, and marks any feature vectors deviating from the normal range as potential anomalies, forming an anomaly event list. Based on this list, anomaly levels are classified, and a control response signal adapted to the network's operating status is generated. Finally, combining the action feedback data after the response signal is executed with the anomaly event handling trajectory, a final monitoring report is output. This system achieves intelligent management and control of underground environmental data from acquisition to processing; it dynamically optimizes sensor configuration based on environmental type classification results, improving the accuracy of data acquisition and the adaptability of sensor installation; it selects low-attenuation transmission paths based on the random forest algorithm, ensuring the stability and integrity of data transmission; and it generates targeted control response signals through anomaly detection and classification, and forms monitoring reports by combining feedback data, enabling rapid identification, accurate classification, and efficient handling of pipeline anomalies. This effectively solves problems such as rigid configuration, transmission distortion, and delayed response in traditional monitoring, providing reliable technical support for the refined operation and maintenance of urban drainage pipeline networks. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A structural block diagram of an Internet of Things-based urban drainage network monitoring and management system provided in an embodiment of the present invention;

[0046] Figure 2 A flowchart illustrating the IoT-based urban drainage network monitoring and management method provided in this embodiment of the invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] First, the implementation environment of the embodiments of this application will be described. Exemplarily, this implementation environment includes key nodes of the urban drainage pipe network, edge computing gateways, cloud server clusters, and pipe network field execution units.

[0049] In the IoT-based urban drainage network monitoring and management system, multimodal sensing devices deployed at key nodes of the network establish connections with edge computing gateways via wireless communication technologies (such as LoRa and NB-IoT) or wired connections, uploading collected underground environmental data and equipment status data to the edge computing gateways in real time. The edge computing gateways achieve bidirectional communication with cloud server clusters via broadband networks, 5G, or industrial Ethernet. On the one hand, they upload the pre-processed and standardized data to the cloud server clusters; on the other hand, they receive sensor configuration adjustment commands and transmission path optimization parameters issued by the cloud server clusters. After the cloud server clusters process the data using core algorithms to generate control response signals, they send them to the network field execution units via the edge computing gateway or directly through the communication link. Simultaneously, the network field execution units transmit feedback data such as action execution status and real-time network operating parameters back to the cloud server clusters, ensuring the real-time performance and stability of data interaction in all aspects of the system.

[0050] Multimodal sensing devices deployed at key nodes of urban drainage pipe networks (such as pipe connection points, low-lying areas, and pipe network intersections) include soil pressure sensors, humidity sensors, and signal transmission status monitoring modules, used to collect core parameters of the underground environment and data related to signal transmission.

[0051] Edge computing gateways are deployed in the communication coverage area around the pipeline network to receive raw data collected by multimodal sensing devices, perform preliminary processing such as format normalization and redundancy removal, and realize bidirectional data transmission with the cloud.

[0052] The cloud server cluster, equipped with support vector machine algorithm models, random forest algorithm models, anomaly detection models, and historical data storage, is used to perform core computing tasks such as underground environment classification, transmission path prediction, anomaly identification, and response signal generation.

[0053] The pipeline field execution unit includes valve controllers, pump group control equipment, and feedback data acquisition modules. It is used to receive control response signals sent by the cloud server cluster and execute corresponding operations, while collecting feedback data such as the completion status of actions and changes in pipeline parameters.

[0054] In one embodiment, such as Figure 1 As shown, this application provides an Internet of Things-based urban drainage network monitoring and management system, which may include:

[0055] The data acquisition module 101 is used to acquire underground environmental data collected by sensors deployed in the urban drainage network to obtain an initial environmental dataset.

[0056] Specifically, underground environmental data is collected by sensors deployed at key nodes of the urban drainage network (such as pipe joints, low-lying areas prone to water accumulation, and pipe network intersections). These sensors may include multimodal sensing devices such as soil pressure sensors and humidity sensors. The collected data covers core indicators such as soil pressure values ​​and soil moisture content, while also recording metadata such as data collection timestamps and sensor deployment location coordinates. This module performs preliminary formatting and redundancy removal on the collected raw data, forming an initial environmental dataset containing environmental parameters, collection information, and location identifiers.

[0057] The configuration optimization module 102 is used to classify underground environment types based on the initial environment dataset using the support vector machine algorithm to obtain environment type labels; it is also used to adjust the sensor installation position and fixing method based on the environment type labels to determine the sensor configuration scheme that is suitable for the current burial depth.

[0058] First, key environmental parameters such as soil pressure are extracted from the initial environmental dataset. These parameters are then input into a trained support vector machine (SVM) algorithm model for underground environment type classification. The classification process incorporates preset soil pressure thresholds and other criteria to output environmental type labels containing environmental characteristic identifiers (e.g., high-pressure environment label, ordinary environment label). Subsequently, based on the underground environmental characteristics corresponding to the environmental type labels (e.g., soil density and pressure transmission characteristics under high pressure), and with the goal of geometric coverage theory and minimizing monitoring blind spots, the module adjusts the initial sensor installation parameters (e.g., spatial distribution and height coordinates of installation points). It also determines suitable sensor fixing method parameters (e.g., fixing stiffness and anchoring depth) by considering the current burial depth of the sensors, soil medium characteristics, and pressure tolerance requirements. Finally, the optimized installation location data and fixing method parameters are coupled and integrated to form a sensor configuration scheme adapted to the current underground geological environment.

[0059] The transmission optimization module 103 is used to predict the transmission path using the random forest algorithm according to the sensor configuration scheme to obtain an optimized transmission path set; if there is a low-attenuation path in the optimized transmission path set, the low-attenuation path is selected for signal transmission to obtain a stable transmission data stream.

[0060] Furthermore, based on the sensor installation location, burial depth, and underground environment characteristics in the sensor configuration scheme, signal transmission-related feature parameters (such as sensor deployment coordinates, soil medium attenuation coefficient, and path occlusion) are extracted and input into the random forest algorithm model to predict suitable signal transmission paths, generating an optimized transmission path set containing multiple candidate paths and attenuation parameters for each path. This module performs attenuation quantification analysis on each path in the set to determine whether there are low-attenuation paths with attenuation below a preset threshold. If so, the low-attenuation path with the smallest attenuation is selected first for signal transmission. During transmission, measures such as terminal impedance matching can be used to reduce signal reflection and crosstalk, ultimately obtaining a stable transmission data stream with high data integrity and low distortion.

[0061] The anomaly handling module 104 is used to analyze the feature vectors in the stable transmission data stream using an anomaly detection model. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined. It is also used to classify the anomaly level based on the list of anomalies and generate control response signals adapted to the pipeline network operation status. Furthermore, it is used to obtain the final monitoring report based on the action feedback data after the response signals are executed and the anomaly event handling trajectory.

[0062] First, feature vectors containing static parameters such as soil pressure and humidity, and dynamic parameters such as signal transmission jitter and attenuation fluctuation rate, are extracted from the stable data stream. An anomaly detection model trained on historical normal data streams is used to analyze these feature vectors. A dual-dimensional judgment rule is introduced, considering both parameter deviation amplitude and deviation rate. If a feature vector exceeds the normal threshold range or the deviation rate exceeds a preset threshold, it is marked as a potential anomaly. Information such as the anomaly occurrence time, pipeline topology location, parameter deviation degree, and associated monitoring nodes are recorded and integrated to form an anomaly event list. Subsequently, based on the parameter deviation degree, pipeline key area weight, and predicted impact range of each anomaly in the list, the analytic hierarchy process (AHP) is used to classify anomalies into levels. The anomaly level, location, and type are then correlated with real-time pipeline operating parameters and a historical database of similar anomaly handling cases to generate differentiated control response signals containing execution priority, target execution unit, and parameter adjustment thresholds. Finally, the module acquires action feedback data such as the action completion degree and parameter callback curve after the control response signal is issued and executed. Combined with the entire chain processing trajectory of abnormal events from marking, level classification, response generation to execution feedback, it integrates information such as abnormal root cause analysis, differentiated response measures and working condition improvement effects, and outputs the final monitoring report.

[0063] The aforementioned IoT-based urban drainage network monitoring and management system comprises the following modules: a data acquisition module obtains underground environmental data collected by sensors deployed in the urban drainage network to form an initial environmental dataset; a configuration optimization module uses this initial environmental dataset as input, employs a support vector machine algorithm to classify underground environmental types, outputs environmental type labels, and then adjusts the sensor installation positions and fixing methods based on these labels to determine a sensor configuration scheme suitable for the current burial depth; a transmission optimization module predicts signal transmission paths using a random forest algorithm based on the sensor configuration scheme, generates an optimized transmission path set, and selects a low-attenuation path for signal transmission if a low-attenuation path exists in the optimized transmission path set, resulting in a stable transmission data stream; an anomaly handling module extracts feature vectors from the stable transmission data stream, analyzes these feature vectors using an anomaly detection model, marks feature vectors as potential anomalies if they deviate from the normal range, and forms an anomaly event list, then classifies anomalies based on the anomaly event list, generates control response signals adapted to the network's operating status, and finally outputs a final monitoring report by combining the action feedback data after the response signals are executed with the anomaly event handling trajectory. This system achieves intelligent management and control of underground environmental data from acquisition to processing; it dynamically optimizes sensor configuration based on environmental type classification results, improving the accuracy of data acquisition and the adaptability of sensor installation; it selects low-attenuation transmission paths based on the random forest algorithm, ensuring the stability and integrity of data transmission; and it generates targeted control response signals through anomaly detection and classification, and forms monitoring reports by combining feedback data, enabling rapid identification, accurate classification, and efficient handling of pipeline anomalies. This effectively solves problems such as rigid configuration, transmission distortion, and delayed response in traditional monitoring, providing reliable technical support for the refined operation and maintenance of urban drainage pipeline networks.

[0064] In one embodiment, the configuration optimization module can also be used for:

[0065] Step S101: Extract soil pressure data from the initial environmental dataset, and use the support vector machine algorithm to classify the underground environment type based on the soil pressure data to obtain the underground environment classification result.

[0066] Step S102: Based on the underground environment classification results and the preset soil pressure threshold, the soil pressure data is judged. If the soil pressure data exceeds the preset threshold, the corresponding underground environment is classified as a high-pressure environment, and an environment type label with environmental characteristic identifier is generated.

[0067] Step S103: Based on the underground environment characteristics corresponding to the environment type label, adjust the initial installation parameters of the sensor according to the geometric coverage theory and the goal of minimizing the monitoring blind zone, and generate optimized sensor installation location data.

[0068] Step S104: Based on the optimized sensor installation location data, combined with the current burial depth of the sensor, soil medium characteristics, and pressure tolerance requirements, the appropriate sensor fixing method parameters are calculated through a multi-objective optimization function.

[0069] Step S105: The optimized sensor installation location data and sensor fixing method parameters are coupled and integrated to obtain a sensor configuration scheme adapted to the underground geological environment.

[0070] First, soil pressure data is extracted from the initial environmental dataset. This data is used as the core input, and a support vector machine algorithm is employed to classify the subsurface environment types, yielding classification results. Then, based on these classification results and a preset soil pressure threshold, the extracted soil pressure data is re-evaluated. If the soil pressure data exceeds the threshold, the corresponding area's subsurface environment is classified as a high-pressure environment, and an environmental type label with accompanying environmental characteristic identifiers (including key information such as soil density and pressure conduction characteristics) is generated. Next, based on the subsurface environment characteristics represented by the environmental type label, and guided by geometric cover theory with the goal of minimizing monitoring blind spots, adjustments are made... The initial sensor installation parameters (including the spatial distribution of installation points and their coordinates relative to the pipeline network segments) are used to generate optimized sensor installation location data. Then, based on this optimized data, the current burial depth of the sensor, soil medium characteristics (such as soil internal friction angle, cohesion, and equivalent stiffness), and pressure tolerance requirements are integrated. A multi-objective optimization function is then used to calculate suitable sensor fixing method parameters (such as anchoring depth, fastener stiffness, protective layer thickness, and fixing angle). Finally, the optimized sensor installation location data and the calculated sensor fixing method parameters are coupled and integrated to form a sensor configuration scheme adapted to the current underground geological environment.

[0071] This embodiment achieves precise adaptation of sensor configuration to the characteristics of the underground environment. By combining support vector machine (SVM) classification with a preset threshold for secondary judgment, the accuracy of underground environment type identification is improved, ensuring the targeted identification of high-pressure environments. Installation parameters are adjusted based on geometric coverage theory and the goal of minimizing monitoring blind spots, effectively reducing monitoring blind spots and ensuring comprehensive data acquisition. Fixing parameters are calculated using a multi-objective optimization function, considering both sensor installation stability and environmental adaptability, taking into account multi-dimensional environmental and equipment factors. Finally, the sensor configuration scheme formed through parameter coupling and integration can adapt to different complex underground geological environments, especially meeting the installation requirements under high-pressure environments, improving the reliability of sensor operation and the effectiveness of data acquisition, and providing high-quality basic data support for subsequent transmission optimization and anomaly detection.

[0072] In one embodiment, the multi-objective optimization function is expressed using the following formula:

[0073]

[0074] in, Represents the vector of optimization variables. , Indicates the anchoring depth. Indicates the stiffness of the fastener. Indicates the thickness of the protective layer. Indicates a fixed angle. , , Represents dynamic weighting coefficients based on environment type labels. Dynamic allocation Describe the objective function. , Indicates the upper limit of allowable displacement. , Indicates soil pressure, Indicates the contact area of ​​the fastener. Indicates the internal friction angle of the soil. This represents soil cohesion, set based on pressure tolerance requirements. Describe the objective function. , This indicates the maximum allowable adaptation deviation. , Indicates the current burial depth of the sensor. Indicates the dynamic adaptation coefficient of anchoring depth. Indicates the equivalent stiffness of the soil. , This indicates the optimal thickness of the protective layer. / 8, Indicates the wavelength of the sensor signal. Describe the objective function. , Indicates the maximum allowable attenuation of the signal. , This indicates the attenuation coefficient of the protective layer material, calibrated in conjunction with the characteristics of the soil medium. , This represents dynamic constraints. , , This represents the combined stress on the sensor's mounting structure under soil pressure. , This represents the normal stress borne by a fixed structure. , This indicates the cross-sectional area of ​​the fixed component under stress. This represents the shear stress borne by a fixed structure. , This indicates the allowable stress of the fastener material.

[0075] Preferably, environment type label For binary variables, =1 corresponds to a high-pressure environment. =0 corresponds to a normal environment, directly assigned by the previous subsurface environment classification results; dynamic weight coefficient. , , satisfy + + =1, high-pressure environment ( =1) down ≥0.5、 ≤0.3、 ≤0.2, normal environment ( =0) ≥0.4 ≤0.4、 ≤0.2, the specific value can be determined by combining the analytic hierarchy process with engineering experience. The value is set according to the sensor's pressure tolerance level, and the range is 0.5-2mm; The value was fixed at 0.8, and was calibrated using adaptation tests of three typical soil environments (sandy soil, clay soil, and silty soil). The value is set based on the sensor signal transmission requirements and should not exceed 5dB. = / 10、 = / 4, =30° =90°; The attenuation coefficient of the protective layer material ranges from 1.0 to 1.5, with 1.5 for metal protective layers and 1.0 for polymer protective layers, calibrated through signal attenuation tests.

[0076] This embodiment's multi-objective optimization function achieves accurate solution for sensor fixing method parameters by fusing three-dimensional objectives: stability, adaptability, and interference-free operation. This solves the problems of poor adaptability and insufficient stability caused by the single-objective orientation in traditional fixing method design. The function introduces dynamic weight coefficients and dynamic constraints, which can be adjusted based on environment type labels. It adapts in real time to high-pressure or ordinary underground environments, ensuring the priority matching of optimization objectives under different working conditions. Simultaneously, it utilizes comprehensive stress constraints. Ensuring the mechanical safety of the fixed structure and preventing structural deformation or damage due to soil pressure, the final output fixing method parameters, after being coupled with the installation location data, can significantly improve the sensor's installation stability, environmental adaptability, and signal transmission integrity in complex underground environments.

[0077] In one embodiment, the transmission optimization module can also be used for:

[0078] Step S201: Obtain the real-time environmental data acquisition results under the sensor configuration scheme.

[0079] Step S202: Input the signal strength, soil medium attenuation coefficient, and path occlusion feature parameters from the acquisition results into the random forest algorithm to predict the appropriate signal transmission path and obtain the optimized transmission path set.

[0080] Step S203: The scattering parameter method is used to quantitatively analyze the dissipation attenuation and reflection attenuation of each path in the optimized transmission path set, and to determine whether there are low attenuation paths with attenuation below the preset signal attenuation threshold.

[0081] Step S204: If a low-attenuation path exists in the optimized transmission path set, the low-attenuation path with the smallest attenuation is selected first for signal transmission. By matching the terminal impedance, signal reflection and crosstalk are reduced, and a stable transmission data stream is obtained.

[0082] Specifically, the system acquires real-time environmental data collected by sensors under a given sensor configuration scheme. Signal strength, soil medium attenuation coefficient, and path occlusion characteristic parameters (including occlusion type, occlusion area, and distribution location) are extracted from the real-time environmental data acquisition results. These parameters are then used as input variables and substituted into a trained random forest algorithm model. The model predicts the adaptability of different transmission paths and outputs an optimized transmission path set containing multiple candidate paths and their basic characteristics. The scattering parameter method is used to perform attenuation quantification analysis on each candidate path in the optimized transmission path set. The dissipation attenuation (signal loss due to soil medium absorption and scattering) and reflection attenuation (signal reflection loss due to path interface impedance mismatch) are calculated for each path. The sum of these two attenuation values ​​is compared with a preset signal attenuation threshold to determine if there are low-attenuation paths in the set with a summed attenuation value lower than the preset threshold. If a low-attenuation path is found, the path with the smallest summed attenuation value is selected as the optimal transmission path. During signal transmission, terminal impedance matching technology is used to adjust the transmission link impedance, reducing signal reflection and crosstalk, ultimately resulting in a stable transmission data stream with low data distortion and high transmission stability.

[0083] This embodiment achieves precise adaptation of signal transmission paths and effective assurance of transmission quality. By introducing the random forest algorithm and combining it with environmental and equipment parameters associated with the sensor configuration scheme to predict the transmission path, the targeting and scientific nature of path selection are improved. The scattering parametric method is used to quantify and analyze two attenuation values, which is more comprehensive than a single attenuation assessment and ensures the accuracy of low-attenuation path identification. At the same time, by prioritizing the path with the smallest attenuation and combining it with terminal impedance matching technology, the interference of factors such as soil medium and path obstruction in complex underground environments on signal transmission is effectively reduced. This solves the problems of blind path selection, large signal loss, and severe data distortion in traditional transmission methods. The stable output transmission data stream provides high-quality data input for the subsequent anomaly detection module, ensuring the accuracy of anomaly identification and analysis and supporting the refined implementation of drainage network monitoring.

[0084] In one embodiment, the exception handling module may further include:

[0085] Anomaly detection unit, used for:

[0086] Step S301: Extract static parameters including soil pressure and humidity, as well as feature vectors of signal transmission jitter rate and attenuation fluctuation rate from the stable transmission data stream.

[0087] Step S302: The anomaly detection model trained based on historical normal data stream is used to analyze the feature vector. If the feature vector exceeds the normal threshold range or the deviation rate exceeds the preset threshold, it is marked as a potential anomaly.

[0088] Step S303: Record the occurrence time, pipeline topology location, parameter deviation magnitude, and associated monitoring nodes of potential anomalies to determine the list of abnormal events.

[0089] The exception handling unit is used for:

[0090] Step S304: Based on the parameter deviation amplitude of each anomaly in the abnormal event list, the weight of the key area of ​​the pipeline network and the range of impact prediction are fused, and the analytic hierarchy process is used to classify the anomaly level.

[0091] Step S305 involves performing correlation analysis on the abnormal event's abnormality level, topology location, parameter type, real-time pipeline operation parameters, and historical similar abnormality handling case library to generate differentiated control response signals, including execution priority, target execution unit, and parameter adjustment threshold.

[0092] Step S306: Obtain action feedback data after the control response signal is issued and executed, combine it with the handling trajectory of abnormal events, integrate the root cause analysis of abnormalities, differentiated response measures and the effect of improving working conditions, and obtain the final monitoring report.

[0093] Schematic, the anomaly identification unit takes a stable transmission data stream as input and extracts two types of core parameters to construct a feature vector: one type is static environmental parameters such as soil pressure and humidity, and the other type is dynamic transmission parameters such as signal transmission jitter rate and attenuation fluctuation rate. This feature vector is input into an anomaly detection model trained based on historical normal data streams. The model adopts a two-dimensional judgment rule of "feature vector numerical range + parameter deviation rate". If the feature vector value exceeds the preset normal threshold range, or the parameter deviation rate exceeds the preset rate threshold, it is marked as a potential anomaly. Then, the key information of the potential anomaly is recorded, including the anomaly occurrence timestamp, the corresponding pipeline topology location coordinates, the parameter deviation amplitude (the difference between the feature vector and the normal benchmark value), and the associated monitoring node number. The above information is structured and integrated to form an anomaly event list. The anomaly handling unit uses the list of anomalies as its core basis, combines the quantified values ​​of parameter deviation amplitudes for each anomaly, integrates the dynamic weights of key areas of the pipeline network (adjusted according to real-time load) and the predicted impact range (predicted through the pipeline hydraulic model), and uses the analytic hierarchy process (AHP) to classify anomalies into different levels. Furthermore, it performs multi-dimensional correlation analysis on anomaly levels, topological locations, parameter types, real-time pipeline operating parameters (such as flow rate and liquid level), and a historical database of similar anomaly handling cases to generate differentiated control response signals containing execution priority, target execution unit address, and parameter adjustment thresholds. Finally, it obtains feedback data such as the completion rate of actions and parameter callback curves after the control response signal is sent to the field execution unit. Combining the entire processing trajectory of anomalies from identification, marking, and level classification to response execution, it integrates the anomaly root cause analysis results, details of differentiated response measures, and the pipeline operating condition improvement effects to form the final monitoring report.

[0094] This embodiment achieves refined end-to-end management and control of pipeline anomalies from discovery to handling. In the anomaly identification phase, a feature vector combining static and dynamic parameters and a two-dimensional judgment rule are employed to improve the accuracy and comprehensiveness of potential anomaly identification, avoiding missed or false judgments caused by single threshold judgments. In the anomaly handling phase, multi-dimensional factors are combined to classify levels and correlate with historical cases to ensure the targeted and scientific nature of control response signals. The differentiated design can adapt to the handling needs of different anomaly scenarios. Simultaneously, through the integrated analysis of the entire process trajectory and action feedback data, quantitative evaluation and root cause tracing of anomaly handling effects are achieved. This solves the problems of one-sided identification, rigid response, and lack of feedback in traditional anomaly handling, providing accurate anomaly information and optimization suggestions for pipeline network operation and maintenance, ensuring the stable and controllable operation of the drainage pipeline network, and supporting the scientific implementation of operation and maintenance decisions.

[0095] In one embodiment, such as Figure 2 As shown, this application also provides a method for monitoring and managing urban drainage networks based on the Internet of Things (IoT), which may include the following steps:

[0096] Step S401: Obtain underground environmental data collected by sensors deployed in the urban drainage network to obtain the initial environmental dataset.

[0097] Step S402: Based on the initial environmental dataset, the support vector machine algorithm is used to classify the underground environment types and obtain the environment type labels.

[0098] Step S403: Adjust the sensor installation location and fixing method based on the environment type label to determine the sensor configuration scheme that is suitable for the current burial depth.

[0099] Step S404: Based on the sensor configuration scheme, the random forest algorithm is used to predict the transmission path to obtain an optimized transmission path set.

[0100] Step S405: If a low-attenuation path exists in the optimized transmission path set, then the low-attenuation path is selected for signal transmission to obtain a stable transmission data stream.

[0101] Step S406: The anomaly detection model is used to analyze the feature vectors in the stable data stream. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined.

[0102] Step S407: Based on the list of abnormal events, classify the abnormality levels and generate control response signals that are adapted to the operating status of the pipeline network.

[0103] Step S408: Based on the action feedback data after the response signal is executed and the abnormal event handling trajectory, the final monitoring report is obtained.

[0104] The aforementioned IoT-based urban drainage network monitoring and management method first acquires underground environmental data collected by sensors deployed at key nodes of the urban drainage network. This data includes core indicators such as soil pressure and humidity, as well as metadata such as collection timestamps and sensor location coordinates. After preliminary formatting and redundancy removal, an initial environmental dataset is obtained. Using this initial environmental dataset as input, key parameters such as soil pressure are extracted. A support vector machine algorithm is used to classify underground environmental types, outputting environmental type labels containing environmental characteristic identifiers. Based on the underground environmental characteristics corresponding to these labels, the sensor installation positions and fixing methods are adjusted to determine a sensor configuration scheme suitable for the current burial depth. According to the sensor configuration scheme, signal transmission-related feature parameters are extracted, and a random forest algorithm is used to predict suitable signal transmission paths, resulting in an optimized transmission path set. Attenuation analysis is performed on each path in this set. If a low-attenuation path with an attenuation amount below a preset threshold exists, this path is selected for signal transmission. Combined with terminal impedance matching technology, signal interference is reduced, resulting in a stable data transmission stream. An anomaly detection model is used to analyze the feature vectors (including static parameters such as soil pressure and humidity, and dynamic parameters such as signal transmission jitter rate and attenuation fluctuation rate) extracted from the stable transmission data stream. If the feature vector deviates from the normal range, it is marked as a potential anomaly. Information such as the time of anomaly occurrence and the location of the pipeline topology is recorded, and an anomaly event list is determined. Based on this list, the anomaly level is classified, and a control response signal adapted to the pipeline network operation status is generated. Finally, the final monitoring report is obtained by combining the action feedback data after the response signal is executed with the anomaly event processing trajectory.

[0105] This embodiment achieves intelligent and refined monitoring and management of urban drainage pipe networks. By using a support vector machine algorithm to classify the underground environment based on initial data, it dynamically optimizes sensor configuration, improving the adaptability and accuracy of data acquisition. The random forest algorithm is used to select low-attenuation transmission paths, combined with impedance matching technology, to ensure the stability and integrity of data transmission. Multi-dimensional feature vectors and anomaly detection models are employed, combined with grading and correlation analysis to generate differentiated response signals, enabling accurate anomaly identification, scientific classification, and efficient handling. The entire process effectively solves problems such as poor data acquisition adaptability, transmission distortion, and delayed anomaly response in traditional monitoring. Through precise control and feedback optimization of the entire data flow, it provides reliable decision-making basis for pipe network operation and maintenance, ensuring the stable and controllable operation of the drainage pipe network.

[0106] It should be understood that although the steps in the flowcharts of the embodiments described above 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 flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0107] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the Internet of Things-based urban drainage network monitoring and management system as described above.

[0108] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0109] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0110] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. An Internet of Things-based urban drainage network monitoring and management system, characterized in that, The system includes: The data acquisition module is used to acquire underground environmental data collected by sensors deployed in the urban drainage network to obtain an initial environmental dataset; The configuration optimization module is used to classify underground environment types based on the initial environment dataset using a support vector machine algorithm to obtain environment type labels; it is also used to adjust the sensor installation position and fixing method based on the environment type labels to determine a sensor configuration scheme suitable for the current burial depth. The transmission optimization module is used to predict the transmission path using a random forest algorithm based on the sensor configuration scheme to obtain an optimized transmission path set; if there is a low-attenuation path in the optimized transmission path set, the low-attenuation path is selected for signal transmission to obtain a stable transmission data stream. The anomaly handling module is used to analyze the feature vectors in the stable transmission data stream using an anomaly detection model. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined. It is also used to classify the anomaly levels based on the list of anomalies and generate control response signals adapted to the pipeline network operation status. Furthermore, it is used to obtain a final monitoring report based on the action feedback data after the response signals are executed and the anomaly event handling trajectory.

2. The system according to claim 1, characterized in that, The configuration optimization module is also used for: Soil pressure data is extracted from the initial environmental dataset, and the subsurface environment type is classified based on the soil pressure data using the support vector machine algorithm to obtain the subsurface environment classification result; Based on the underground environment classification results and a preset soil pressure threshold, the soil pressure data is judged. If the soil pressure data exceeds the preset threshold, the corresponding underground environment is classified as a high-pressure environment, and an environmental type label with environmental characteristic identifier is generated. Based on the underground environment characteristics corresponding to the environmental type label, the initial installation parameters of the sensor are adjusted according to the geometric coverage theory and the goal of minimizing the monitoring blind zone, and optimized sensor installation location data is generated. Based on the optimized sensor installation location data, combined with the current burial depth of the sensor, soil medium characteristics, and pressure tolerance requirements, the appropriate sensor fixing method parameters are calculated through a multi-objective optimization function. The optimized sensor installation location data is coupled and integrated with the sensor fixing method parameters to obtain a sensor configuration scheme adapted to the underground geological environment.

3. The system according to claim 2, characterized in that, The multi-objective optimization function is expressed by the following formula: in, Represents the vector of optimization variables. , Indicates the anchoring depth. Indicates the stiffness of the fastener. Indicates the thickness of the protective layer. Indicates a fixed angle. , , Represents dynamic weighting coefficients based on environment type labels. Dynamic allocation Describe the objective function. , Indicates the upper limit of allowable displacement. , Indicates soil pressure, Indicates the contact area of ​​the fastener. Indicates the internal friction angle of the soil. This represents soil cohesion, set based on pressure tolerance requirements. Describe the objective function. , This indicates the maximum allowable adaptation deviation. , Indicates the current burial depth of the sensor. Indicates the dynamic adaptation coefficient of anchoring depth. Indicates the equivalent stiffness of the soil. , This indicates the optimal thickness of the protective layer. / 8, Indicates the wavelength of the sensor signal. Describe the objective function. , Indicates the maximum allowable attenuation of the signal. , This indicates the attenuation coefficient of the protective layer material, calibrated in conjunction with the characteristics of the soil medium. , This represents dynamic constraints. , , This represents the combined stress on the sensor's mounting structure under soil pressure. , This represents the normal stress borne by a fixed structure. , This indicates the cross-sectional area of ​​the fixed component under stress. This represents the shear stress borne by a fixed structure. , This indicates the allowable stress of the fastener material.

4. The system according to claim 1, characterized in that, The transmission optimization module is also used for: Obtain the real-time environmental data acquisition results under the aforementioned sensor configuration scheme; The signal strength, soil medium attenuation coefficient, and path occlusion characteristic parameters from the collected results are input into the random forest algorithm to predict the appropriate signal transmission path, thereby obtaining an optimized transmission path set. The scattering parameter method is used to quantitatively analyze the dissipation attenuation and reflection attenuation of each path in the optimized transmission path set to determine whether there are low attenuation paths with attenuation below the preset signal attenuation threshold. If the set of optimized transmission paths contains a low-attenuation path, the low-attenuation path with the smallest attenuation is selected first for signal transmission. By reducing signal reflection and crosstalk through terminal impedance matching, a stable transmission data stream is obtained.

5. The system according to claim 1, characterized in that, The exception handling module also includes: Anomaly detection unit, used for: Extract static parameters including soil pressure and humidity, as well as feature vectors of signal transmission jitter rate and attenuation fluctuation rate from the stable transmission data stream. The feature vector is analyzed using an anomaly detection model trained on historical normal data streams. If the feature vector exceeds the normal threshold range or the deviation rate exceeds a preset threshold, it is marked as a potential anomaly. Record the occurrence time, pipeline topology location, parameter deviation magnitude, and associated monitoring nodes of the potential anomalies to determine the list of abnormal events; The exception handling unit is used for: Based on the parameter deviation amplitude of each anomaly in the list of abnormal events, the weight of key areas of the pipeline network and the range of impact prediction are fused, and the analytic hierarchy process is used to classify the anomaly levels. The abnormal event's abnormal level, topological location, parameter type, real-time pipeline operation parameters, and historical similar abnormal handling case library are correlated and analyzed to generate differentiated control response signals including execution priority, target execution unit, and parameter adjustment threshold. The action feedback data after the control response signal is issued and executed is obtained. Combined with the processing trajectory of the abnormal event, the root cause analysis of the abnormality, the differentiated response measures and the effect of the improvement of the working condition are integrated to obtain the final monitoring report.

6. A method for monitoring and managing urban drainage pipe networks based on the Internet of Things, characterized in that, The method includes: Obtain underground environmental data collected by sensors deployed in the urban drainage network to obtain an initial environmental dataset; Based on the initial environmental dataset, the support vector machine algorithm is used to classify the underground environment types and obtain environmental type labels; Adjust the sensor installation position and fixing method based on the environmental type label to determine a sensor configuration scheme suitable for the current burial depth; Based on the sensor configuration scheme, the random forest algorithm is used to predict the transmission path to obtain an optimized transmission path set; If a low-attenuation path exists in the optimized transmission path set, then the low-attenuation path is selected for signal transmission to obtain a stable transmission data stream. An anomaly detection model is used to analyze the feature vectors in the stable transmission data stream. If the feature vectors deviate from the normal range, they are marked as potential anomalies and a list of anomaly events is determined. Based on the list of abnormal events, the abnormality levels are classified to generate control response signals that are adapted to the operating status of the pipeline network. Based on the action feedback data and abnormal event handling trajectory after the response signal is executed, the final monitoring report is obtained.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the system according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the system according to any one of claims 1 to 5.

Images