A sewage pipeline operation and maintenance supervision system and a supervision method thereof

Abstract

The present application provides a sewage pipeline operation and maintenance supervision system and a supervision method thereof, and relates to the technical field of sewage pipeline operation and maintenance supervision. The sewage pipeline operation and maintenance supervision system comprises a sewage pipeline operation and maintenance supervision system, which comprises a sensing layer, a transmission layer, a data processing layer, an application layer and an operation and maintenance execution layer. The sensing layer comprises a multi-parameter fusion sensor group arranged at key nodes of the sewage pipeline, which is used for collecting data of flow, liquid level, pressure, pH value, dissolved oxygen and hydrogen sulfide concentration in the pipeline in real time. The transmission layer adopts a 5G+LoRa dual-channel transmission mode to transmit the data collected by the sensing layer to the data processing layer. The data processing layer comprises a data preprocessing module, a blockage prediction module, a health degree evaluation module and a fault diagnosis module. Through the implementation of multi-dimensional sensing transmission, intelligent analysis and prediction, active early warning and scheduling and full-process closed-loop management, the operation and maintenance efficiency and the intelligent level of the sewage pipeline network are comprehensively improved.

Description

Technical Field

[0001] This invention relates to the field of sewage pipeline operation and maintenance supervision technology, specifically to a sewage pipeline operation and maintenance supervision system and its supervision method. Background Technology

[0002] Urban sewage pipeline systems are an important component of urban infrastructure, playing a crucial role in collecting and transporting domestic sewage and industrial wastewater, and directly impacting urban public health and environmental quality. With the acceleration of urbanization in my country, the scale of urban sewage pipe networks has continuously expanded, with the total length exceeding one million kilometers. The network structure has become increasingly complex, incorporating various materials such as reinforced concrete pipes, ductile iron pipes, and plastic pipes, distributed across different geological conditions and underground spaces.

[0003] However, the existing sewage pipeline operation and maintenance management model still has many technical problems that urgently need to be solved: The operation and maintenance model is passive and lagging behind: it mainly relies on regular manual inspections and public reports to discover problems, and cannot provide early warnings and accurate location of the internal conditions of underground pipe networks. Often, it is only dealt with after serious accidents such as sewage overflows and road collapses occur, resulting in high operation and maintenance costs and significant social impact.

[0004] First, sensing technology has limitations: existing monitoring equipment is mostly single-function sensor, which is difficult to acquire multi-dimensional parameters such as flow rate, liquid level, pressure, water quality, and gas concentration in pipelines at the same time. Moreover, it has poor reliability and short lifespan in harsh environments such as sewage corrosion, high humidity, and low light, and the data transmission stability is insufficient.

[0005] Secondly, data analysis capabilities are insufficient: there is a lack of effective data fusion and intelligent analysis algorithms, making it impossible to fully utilize historical operation and maintenance data and real-time monitoring data for pipeline health status assessment and fault prediction. Existing siltation prediction models are mostly based on single hydraulic parameters and do not comprehensively consider the influence of multiple factors such as pipeline material, service life, geological conditions, and surrounding environment, resulting in low prediction accuracy.

[0006] Third, the allocation of operation and maintenance resources is unreasonable: the lack of a scientific fault priority sorting mechanism and intelligent scheduling system leads to the waste of operation and maintenance resources, high-risk pipelines are not maintained in a timely manner, low-risk pipelines are over-maintained, and the overall operation and maintenance efficiency is low.

[0007] Finally, the system integration is low: the data standards between subsystems are not unified, forming "data silos", making it impossible to achieve closed-loop management of the entire process from data collection, analysis, early warning to operation and maintenance. The application of digital twin technology is not deep enough, and there is a lack of intuitive visualization and simulation capabilities. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a sewage pipeline operation and maintenance monitoring system and its monitoring method, which solves the problems of limitations in sensing technology, insufficient data analysis capabilities, unreasonable allocation of operation and maintenance resources, and low system integration.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a sewage pipeline operation and maintenance monitoring system, comprising a sensing layer, a transmission layer, a data processing layer, an application layer, and an operation and maintenance execution layer. The sensing layer includes a multi-parameter fusion sensor group deployed at key nodes of the sewage pipeline, used to collect real-time data on flow rate, liquid level, pressure, pH value, dissolved oxygen, and hydrogen sulfide concentration in the pipeline. The transmission layer adopts a 5G+LoRa dual-channel transmission mode to transmit the data collected by the sensing layer to the data processing layer; The data processing layer includes a data preprocessing module, a blockage prediction module, a health assessment module, and a fault diagnosis module. The application layer includes a digital twin visualization platform, an intelligent early warning module, and an operation and maintenance scheduling module; the operation and maintenance execution layer includes a mobile operation and maintenance terminal and an intelligent dredging robot. The clogging prediction module calculates the pipeline clogging risk index R based on an improved hydrodynamic model and machine learning algorithm, using the following formula: ; in, For real-time traffic, For design traffic, For real-time liquid level, To design the liquid level, For real-time flow rate, The design flow rate is T, and the service life of the pipeline is T. , , , The weighting coefficients are determined through training with historical data.

[0010] Preferably, the multi-parameter fusion sensor group includes a Doppler flow meter, an ultrasonic level gauge, a pressure sensor, a pH sensor, a dissolved oxygen sensor, a hydrogen sulfide gas sensor, and a vibration sensor. All sensors are encapsulated with IP68 waterproof rating and corrosion-resistant materials, and are powered by built-in lithium batteries with a battery life of no less than 3 years.

[0011] Preferably, the data preprocessing module is used to clean, denoise, normalize, and spatiotemporally register the collected raw data, using an adaptive wavelet transform algorithm to remove environmental noise and equipment noise, and using a sliding window method to detect and repair outliers, thereby constructing a standardized dataset.

[0012] Preferably, the health assessment module comprehensively considers the factors in four dimensions of pipeline structure defects, hydraulic performance, environmental impact, and operation and maintenance history, determines the weights of each factor using the analytic hierarchy process, and calculates the comprehensive health index of the pipeline , and the formula is: ; where, is the structural defect score, is the hydraulic performance score, is the environmental impact score, is the operation and maintenance history score, , are the weight coefficients of each dimension, and .

[0013] Preferably, the fault diagnosis module constructs a fault diagnosis model based on the long short-term memory network (LSTM) and the attention mechanism, can automatically identify various fault types such as pipeline siltation, leakage, corrosion, dislocation, and breakage, and realizes the accurate positioning of the fault point through spatio-temporal correlation analysis, and the positioning error does not exceed 5 meters.

[0014] Preferably, the digital twin visualization platform is constructed based on the GIS geographic information system and three-dimensional modeling technology, can display the topological structure, operation status, fault location, and warning information of the sewage pipe network in real time, supports virtual roaming inside the pipeline and " " scenario simulation, and provides intuitive visual support for operation and maintenance decision-making.

[0015] Preferably, the intelligent warning module divides the warning levels into four levels according to the siltation risk index R and the comprehensive health index H: level 1 warning (R≥0.8 or H≤0.2), level 2 warning (0.6≤R<0.8 or 0.2<H≤0.4), level 3 warning (0.4≤R<0.6 or 0.4<H≤0.6), level 4 warning (R<0.4 or H>0.6), and sends warning information to operation and maintenance personnel through various methods such as text messages, APP push, and sound and light alarms.

[0016] Preferably, the operation and maintenance scheduling module constructs an intelligent scheduling model based on the genetic algorithm, comprehensively considers factors such as fault priority, operation and maintenance personnel location, equipment status, and traffic conditions, automatically generates the optimal operation and maintenance scheduling plan, and realizes the reasonable allocation and efficient utilization of operation and maintenance resources.

[0017] The method for operation and maintenance supervision of sewage pipelines includes the following steps: Step S1: Deploy a multi-parameter fusion sensor group at key nodes of the sewage pipeline to collect pipeline operation data in real time; Step S2: Transmit the collected data to the data processing layer through the 5G+LoRa dual-channel transmission mode; Step S3: The data preprocessing module cleans, denoises, and standardizes the raw data; Step S4: The siltation prediction module calculates the siltation risk index R for each pipe segment, and the health assessment module calculates the comprehensive health index H for each pipe segment; Step S5: The fault diagnosis module identifies the fault type and locates the fault point based on the processed data and calculation results; Step S6: The intelligent early warning module generates early warning information of corresponding levels based on the R and H values ​​and pushes it to the operation and maintenance personnel; Step S7: The operation and maintenance scheduling module automatically generates the optimal operation and maintenance scheduling plan and sends it to the mobile operation and maintenance terminal; Step S8: Maintenance personnel go to the fault location according to the scheduling plan and mobile terminal navigation to handle the problem, and upload the handling results through the mobile terminal after the problem is handled. Step S9: The system updates the pipeline database and digital twin model to form a closed-loop management of the entire process.

[0018] Preferably, the calculation of the clogging risk index R in step S4 further includes a real-time correction step: when a clogging failure actually occurs, the system automatically records the values ​​of various parameters at the time of the failure and updates the weight coefficients α, β, and α using an incremental learning algorithm. , This continuously improves the accuracy of siltation prediction; at the same time, the system regularly recalculates and sorts the comprehensive health index H of all pipe sections to generate an annual operation and maintenance plan to guide the implementation of preventive maintenance work.

[0019] This invention provides a sewage pipeline operation and maintenance monitoring system and method. It has the following beneficial effects: This invention employs a fusion sensing scheme integrating multiple types of sensors, which can simultaneously collect multi-dimensional pipeline operating parameters such as flow rate, liquid level, and pressure. The sensors are IP68 waterproof and corrosion-resistant encapsulated to adapt to harsh working conditions. Combined with 5G+LoRa dual-channel redundant transmission, it effectively solves the problems of existing sensing technologies being single, having poor environmental adaptability, and unstable data transmission.

[0020] This invention constructs a multi-dimensional intelligent data analysis system, completes data preprocessing through algorithms such as adaptive wavelet transform, combines an improved hydrodynamic model with machine learning to calculate siltation risk and health index, and uses an LSTM+attention mechanism to achieve accurate fault identification and location, significantly improving the accuracy of pipeline condition assessment and fault prediction.

[0021] This invention transforms the operation and maintenance mode from passive handling to proactive early warning, establishes a four-level early warning system based on dual indices to promptly push risk information, and uses an intelligent scheduling module based on genetic algorithms to generate the optimal solution by integrating multiple factors, rationally allocate operation and maintenance resources, and avoid high-risk pipeline sections being neglected and low-risk pipeline sections being over-maintained.

[0022] This invention achieves deep integration and closed-loop management of the entire system, and provides visualization and simulation support through a digital twin platform based on GIS and 3D modeling. By breaking down data silos through a complete process of data collection, analysis, early warning, handling, and model self-updating, it significantly improves the overall efficiency and intelligence level of pipeline network operation and maintenance. Attached Figure Description

[0023] Figure 1 This is a flowchart of the data processing and analysis process of the present invention; Figure 2 This is a flowchart illustrating the entire process monitoring method of the present invention; Figure 3 This is a comparison chart of the implementation flow rates of the present invention; Figure 4 This is a trend comparison chart of the monitoring data of the present invention; Figure 5 This is a data analysis comparison chart for the present invention; Figure 6 This is a comparison chart of monitoring point data from the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1:

[0026] like Figure 1 , Figure 3 , Figure 4 , Figure 5 and Figure 6 As shown, this embodiment of the invention provides a sewage pipeline operation and maintenance monitoring system, including a sensing layer, a transmission layer, a data processing layer, an application layer, and an operation and maintenance execution layer: The sensing layer includes a multi-parameter fusion sensor group deployed at key nodes of the sewage pipeline for real-time acquisition of data on flow rate, liquid level, pressure, pH value, dissolved oxygen, and hydrogen sulfide concentration within the pipeline. The multi-parameter fusion sensor group includes a Doppler flow meter, an ultrasonic level meter, a pressure sensor, a pH sensor, a dissolved oxygen sensor, a hydrogen sulfide gas sensor, and a vibration sensor. All sensors are IP68 waterproof and encapsulated in corrosion-resistant materials, powered by a built-in lithium battery with a battery life of no less than 3 years. The transmission layer adopts a 5G+LoRa dual-channel transmission mode to transmit the data collected by the sensing layer to the data processing layer; The data processing layer includes a data preprocessing module, a blockage prediction module, a health assessment module, and a fault diagnosis module. The data preprocessing module cleans, denoises, normalizes, and performs spatiotemporal registration on the collected raw data. It uses an adaptive wavelet transform algorithm to remove environmental and equipment noise, detects and repairs outliers using the sliding window method, and constructs a standardized dataset. The health assessment module comprehensively considers four dimensions: pipeline structural defects, hydraulic performance, environmental impact, and operational history. It uses the analytic hierarchy process (AHP) to determine the weights of each factor and calculates the pipeline's overall health index. The formula is: ; in, Score structural defects. To score hydraulic performance, For environmental impact scoring, Scoring based on operational history, , where is the weight coefficient for each dimension, and ; The fault diagnosis module builds a fault diagnosis model based on Long Short-Term Memory Network (LSTM) and attention mechanism. It can automatically identify various fault types such as pipeline siltation, leakage, corrosion, misalignment, and damage. It can also achieve accurate location of fault points through spatiotemporal correlation analysis with a positioning error of no more than 5 meters. The application layer includes a digital twin visualization platform, an intelligent early warning module, and an operation and maintenance scheduling module; the operation and maintenance execution layer includes a mobile operation and maintenance terminal and an intelligent dredging robot. The clogging prediction module calculates the pipeline clogging risk index R based on an improved hydrodynamic model and machine learning algorithm, using the following formula: ; in, For real-time traffic, For design traffic, For real-time liquid level, To design the liquid level, For real-time flow rate, The design flow rate is T, and the service life of the pipeline is T. , , , The weighting coefficients are determined through training with historical data.

[0027] The digital twin visualization platform, built upon GIS geographic information systems and 3D modeling technology, can display the topology, operational status, fault locations, and early warning information of sewage pipe networks in real time, supporting virtual roaming within the pipelines and... "Scenario simulation provides intuitive visualization support for operation and maintenance decisions."

[0028] The intelligent warning module classifies the warning levels into four levels according to the clogging risk index R and the comprehensive health index H: level 1 warning (R≥0.8 or H≤0.2), level 2 warning (0.6≤R<0.8 or 0.2<H≤0.4), level 3 warning (0.4≤R<0.6 or 0.4<H≤0.6), level 4 warning (R<0.4 or H>0.6), and sends warning information to the operation and maintenance personnel through various methods such as text messages, APP push, sound and light alarms; The operation and maintenance scheduling module constructs an intelligent scheduling model based on the genetic algorithm, comprehensively considers factors such as fault priority, location of operation and maintenance personnel, equipment status and traffic conditions, automatically generates the optimal operation and maintenance scheduling plan, and realizes the reasonable allocation and efficient utilization of operation and maintenance resources.

[0029] Specifically: The sewage pipeline operation and maintenance supervision system is composed of a perception layer, a transmission layer, a data processing layer, an application layer and an operation and maintenance execution layer connected in sequence. The multi-parameter fusion sensor group in the perception layer is deployed at key nodes such as inspection wells, branch pipe intersections, pump station entrances and low-lying areas of the sewage pipeline at an interval of every 500 meters. Each group of sensors integrates a Doppler flowmeter, an ultrasonic level gauge, a pressure sensor, a pH sensor, a dissolved oxygen sensor, a hydrogen sulfide gas sensor and a vibration sensor. All sensors adopt a 316L stainless steel shell and an epoxy resin encapsulation process to achieve an IP68 waterproof level, are内置 with a 10Ah lithium iron phosphate battery, and adopt a sleep-wake working mode. Under normal conditions, data is collected once every 15 minutes, and in abnormal conditions, it automatically switches to collect once every 1 minute, and the battery life can reach more than 3 years. The transmission layer adopts a 5G+LoRa dual-channel redundant transmission architecture. The LoRa module is responsible for the low-power long-distance transmission of conventional monitoring data, and the coverage radius can reach 3 kilometers. The 5G module is responsible for the high-speed transmission of large amounts of data such as high-definition videos and fault alarms. When one of the links is interrupted, the system automatically switches to the other link to ensure the continuity and reliability of data transmission; The data processing layer is deployed on a cloud server. The data preprocessing module first uses an adaptive wavelet transform algorithm to decompose and denoise the raw data at five levels. Then, it uses the 3σ criterion combined with the sliding window method to detect outliers and repairs them using the weighted average of data from adjacent time periods. Finally, all data is normalized to the [0,1] interval and spatiotemporally registered to construct a unified format pipeline operation database. The siltation prediction module, based on the improved Saint-Venant hydrodynamic model and random forest algorithm, combined with basic parameters such as pipeline material, diameter, and slope, calculates the siltation risk index R of each pipe section in real time. The weight coefficients α, β, γ, and δ are determined through training with historical operation and maintenance data from the past 5 years, with initial values ​​of 0.3, 0.4, 0.2, and 0.1, respectively. The health assessment module uses the analytic hierarchy process to subdivide the four dimensions of pipeline structural defects, hydraulic performance, environmental impact, and operation and maintenance history into 16 secondary indicators. The weight of each indicator is determined by expert scoring, and a comprehensive health index H between 0 and 1 is calculated. The fault diagnosis module is built based on a bidirectional LSTM network combined with an attention mechanism. It takes pre-processed multi-dimensional time series data as input and outputs the probability of five common faults such as pipeline siltation, leakage, corrosion, misalignment and damage. It also achieves accurate fault location through spatiotemporal correlation analysis of adjacent sensor data, with the location error controlled within 5 meters. The application-layer digital twin visualization platform is built on the ArcGIS geographic information system and the Unity3D engine. It imports pipeline CAD drawings and BIM models to generate a 1:1 scale 3D digital twin, mapping the pipeline network's operational status in real time. It supports 360-degree virtual navigation within the pipeline and hydraulic simulation under different operating conditions. The intelligent early warning module automatically classifies warning levels into four levels based on the blockage risk index R and the comprehensive health index H. Level 1 warnings are simultaneously communicated to the operations and maintenance (O&M) manager via audible and visual alarms, SMS, and telephone. Levels 2 through 4 warnings are pushed to O&M personnel in the corresponding areas via an app. The O&M scheduling module, based on an improved genetic algorithm, uses the highest fault priority and shortest total travel distance as its objective function. It comprehensively considers the real-time location, skill level, equipment status, and real-time traffic conditions of O&M personnel to generate the optimal O&M scheduling plan within 30 seconds. After receiving the task, the mobile operation and maintenance terminal at the operation and maintenance execution layer provides navigation routes and fault details. The intelligent dredging robot can enter the pipeline through remote control to carry out inspection and dredging operations. After the operation is completed, the system automatically uploads the disposal report and on-site photos, updates the pipeline database and digital twin model, and forms a complete closed-loop management process.

[0030] Example 2:

[0031] like Figure 2 As shown, the sewage pipeline operation and maintenance supervision method includes the following steps: Step S1: Deploy multi-parameter fusion sensor groups at key nodes of the sewage pipeline to collect pipeline operation data in real time; Step S2: Transmit the collected data to the data processing layer via 5G+LoRa dual-channel transmission mode; Step S3: The data preprocessing module cleans, denoises, and standardizes the raw data; Step S4: The siltation prediction module calculates the siltation risk index R for each pipe segment, and the health assessment module calculates the comprehensive health index H for each pipe segment. The calculation of the siltation risk index R also includes a real-time correction step: when a siltation fault actually occurs, the system automatically records the values ​​of various parameters at the time of the fault and updates the weight coefficients α, β, and α through an incremental learning algorithm. , The system continuously improves the accuracy of siltation prediction; at the same time, it regularly recalculates and sorts the comprehensive health index H of all pipe sections to generate an annual operation and maintenance plan to guide the implementation of preventive maintenance work. Step S5: The fault diagnosis module identifies the fault type and locates the fault point based on the processed data and calculation results; Step S6: The intelligent early warning module generates early warning information of corresponding levels based on the R and H values ​​and pushes it to the operation and maintenance personnel; Step S7: The operation and maintenance scheduling module automatically generates the optimal operation and maintenance scheduling plan and sends it to the mobile operation and maintenance terminal; Step S8: Maintenance personnel go to the fault location according to the scheduling plan and mobile terminal navigation to handle the problem, and upload the handling results through the mobile terminal after the problem is handled. Step S9: The system updates the pipeline database and digital twin model to form a closed-loop management of the entire process.

[0032] Specifically: First, multi-parameter fusion sensor groups are deployed at key nodes such as inspection wells, branch pipe junctions, pump station inlets and outlets, low-lying areas prone to water accumulation, and old pipe sections with a service life exceeding 15 years, at intervals of 300-500 meters. The sensors adopt a low-power operating mode with alternating sleep and wake-up cycles. Under normal conditions, data on flow rate, liquid level, pressure, pH value, dissolved oxygen, hydrogen sulfide concentration, and vibration within the pipeline are collected every 15 minutes. When any parameter exceeds a preset threshold, the system automatically switches to a 1-minute high-frequency acquisition mode to ensure complete capture of abnormal data. The collected data is uploaded via a 5G+LoRa dual-channel redundant transmission mode. Before transmission, the data is encrypted using the AES-256 algorithm. The LoRa link is responsible for long-distance, low-power transmission of regular small batches of data, while the 5G link is responsible for high-speed transmission of abnormal alarms and high-definition video data. When the signal strength of either link drops below -100dBm, the system automatically switches to the backup link to ensure the continuity of data transmission.

[0033] After the data reaches the cloud data processing layer, the data preprocessing module first uses a 5-level adaptive wavelet transform algorithm to remove environmental noise and equipment electromagnetic interference from the original data. Then, it uses a sliding window method with a window size of 10 combined with the 3σ criterion to detect outliers and repairs them using the weighted average of three adjacent valid data points. Subsequently, all parameters are normalized to the [0,1] interval and spatiotemporal registration is performed to unify them into a standardized time-series dataset with a 1-minute time granularity. Afterward, the blockage prediction module calculates the blockage risk index R of each pipe segment every hour, and the health assessment module recalculates the comprehensive health index H of all pipe segments every month. When a blockage fault actually occurs and is dealt with, the system automatically records the parameters at the time of the fault and the handling results. The weight coefficients α, β, γ, and δ are iteratively updated using an incremental learning algorithm. At the end of each year, all pipe segments are ranked by risk based on the annual H value calculation results, and a preventive maintenance plan for the following year is generated, prioritizing the detection and maintenance of high-risk pipe segments with a comprehensive health index H≤0.4.

[0034] The fault diagnosis module takes into account nearly 24 hours of pre-processed multi-dimensional time-series data and identifies five common fault types—pipeline siltation, leakage, corrosion, misalignment, and damage—based on a bidirectional LSTM-attention mechanism model, achieving an accuracy rate of no less than 95%. It also calculates the precise location of the fault point through spatiotemporal correlation analysis of adjacent sensor data, with a positioning error controlled within 5 meters. The intelligent early warning module automatically classifies four warning levels based on the calculated R and H values. Level 1 warnings simultaneously notify the maintenance supervisor and emergency response team via audible and visual alarms, SMS, and telephone. Level 2 warnings are pushed to the area maintenance manager, and Levels 3 and 4 warnings are pushed to ordinary maintenance personnel in the corresponding grid. Warning information includes the pipe section number, precise location, risk level, fault type, and suggested handling timeframe. The maintenance scheduling module uses the shortest total response time and highest maintenance resource utilization as its objective function. Based on an improved genetic algorithm, it comprehensively considers fault priority, real-time location of maintenance personnel, skill matching, equipment status, and real-time traffic conditions, generating the optimal scheduling plan within 30 seconds and sending it to the mobile maintenance terminal.

[0035] After receiving the task, maintenance personnel use the navigation function of their mobile devices to navigate to the fault location. On-site, they can access the digital twin platform to view historical pipeline operation data and internal structure, use portable testing equipment to verify the fault situation, and upload work progress in real time during the handling process. For complex problems, remote video consultations can be initiated. After the handling is completed, on-site photos, videos, and a detailed handling report are uploaded, and the system automatically conducts acceptance and review. Finally, the system updates the pipe segment health status, fault records, and handling results in the pipeline network database, synchronously updates the operating parameters of the digital twin model, and adds the fault data to the model training set, completing a closed-loop management process from data collection, analysis and early warning, scheduling and handling to model optimization.

[0036] Example 3:

[0037] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A sewage pipeline operation and maintenance supervision system, including a perception layer, a transmission layer, a data processing layer, an application layer and an operation and maintenance execution layer, characterized in that: The perception layer includes a multi-parameter fusion sensor group deployed at key nodes of the sewage pipeline, which is used to collect data such as flow rate, liquid level, pressure, pH value, dissolved oxygen, and hydrogen sulfide concentration in the pipeline in real time; The transmission layer adopts a 5G+LoRa dual-channel transmission mode to transmit the data collected by the perception layer to the data processing layer; The data processing layer includes a data preprocessing module, a clogging prediction module, a health assessment module and a fault diagnosis module; The application layer includes a digital twin visualization platform, an intelligent warning module and an operation and maintenance scheduling module; the operation and maintenance execution layer includes a mobile operation and maintenance terminal and an intelligent dredging robot; The clogging prediction module calculates the pipeline clogging risk index R based on an improved hydrodynamic model and a machine learning algorithm. The formula is: ; in, For real-time traffic, For design traffic, For real-time liquid level, To design the liquid level, For real-time flow rate, The design flow rate is T, and the service life of the pipeline is T. , , , The weighting coefficients are determined through training with historical data.

2. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The multi-parameter fusion sensor group includes a Doppler flowmeter, an ultrasonic level gauge, a pressure sensor, a pH sensor, a dissolved oxygen sensor, a hydrogen sulfide gas sensor and a vibration sensor. All sensors are encapsulated with an IP68 waterproof rating and corrosion-resistant materials, and are powered by an internal lithium battery with a battery life of no less than 3 years.

3. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The data preprocessing module is used to clean, denoise, normalize and spatio-temporally register the collected raw data. It uses an adaptive wavelet transform algorithm to remove environmental noise and equipment noise, detects and repairs outliers through a sliding window method, and constructs a standardized data set.

4. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The health assessment module comprehensively considers four dimensions of factors: pipeline structural defects, hydraulic performance, environmental impact, and operation and maintenance history. It uses the analytic hierarchy process (AHP) to determine the weight of each factor and calculates the overall pipeline health index. The formula is: ; in, Score structural defects. To score hydraulic performance, For environmental impact scoring, Scoring based on operational history, , where is the weight coefficient for each dimension, and .

5. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The fault diagnosis module constructs a fault diagnosis model based on a long short-term memory network (LSTM) and an attention mechanism, which can automatically identify various fault types such as pipeline siltation, leakage, corrosion, dislocation, and damage, and accurately locate the fault point through spatio-temporal correlation analysis, with a positioning error not exceeding 5 meters.

6. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The digital twin visualization platform, built upon GIS geographic information system and 3D modeling technology, can display the topology, operating status, fault location, and early warning information of the sewage pipe network in real time, and supports virtual roaming inside the pipeline and... "Scenario simulation provides intuitive visualization support for operation and maintenance decisions." 7. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The intelligent warning module divides the warning level into four levels according to the clogging risk index R and the comprehensive health index H: level 1 warning (R≥0.8 or H≤0.2), level 2 warning (0.6≤R<0.8 or 0.2<H≤0.4), level 3 warning (0.4≤R<0.6 or 0.4<H≤0.6), level 4 warning (R<0.4 or H>0.6), and sends warning information to operation and maintenance personnel through various methods such as text messages, APP push, and audible and visual alarms.

8. The sewage pipeline operation and maintenance monitoring system according to claim 1, characterized in that: The operation and maintenance scheduling module constructs an intelligent scheduling model based on a genetic algorithm, comprehensively considers factors such as fault priority, operation and maintenance personnel location, equipment status and traffic conditions, automatically generates an optimal operation and maintenance scheduling plan, and realizes the reasonable allocation and efficient utilization of operation and maintenance resources.

9. A method for the operation and maintenance supervision of sewage pipelines, characterized in that, Applied to the sewage pipeline operation and maintenance supervision system according to any one of claims 1 to 8, it includes the following steps: Step S1: Deploy a multi-parameter fusion sensor group at key nodes of the sewage pipeline to collect pipeline operation data in real time; Step S2: Transmit the collected data to the data processing layer through a 5G+LoRa dual-channel transmission mode; Step S3: The data preprocessing module performs cleaning, denoising and standardization processing on the raw data; Step S4: The clogging prediction module calculates the clogging risk index R of each pipe section, and the health assessment module calculates the comprehensive health index H of each pipe section; Step S5: The fault diagnosis module identifies the fault type and locates the fault point based on the processed data and calculation results; Step S6: The intelligent early warning module generates early warning information of corresponding levels based on the R and H values ​​and pushes it to the operation and maintenance personnel; Step S7: The operation and maintenance scheduling module automatically generates the optimal operation and maintenance scheduling plan and sends it to the mobile operation and maintenance terminal; Step S8: Maintenance personnel go to the fault location according to the scheduling plan and mobile terminal navigation to handle the problem, and upload the handling results through the mobile terminal after the problem is handled. Step S9: The system updates the pipeline database and digital twin model to form a closed-loop management of the entire process.

10. The sewage pipeline operation and maintenance supervision method according to claim 9, characterized in that: The calculation of the clogging risk index R in step S4 also includes a real-time correction step: when a clogging failure actually occurs, the system automatically records the values ​​of various parameters at the time of the failure and updates the weight coefficients α, β, and α using an incremental learning algorithm. , This continuously improves the accuracy of siltation prediction; at the same time, the system regularly recalculates and sorts the comprehensive health index H of all pipe sections to generate an annual operation and maintenance plan to guide the implementation of preventive maintenance work.

Images