Method for evaluating upgrading and efficiency effect of sewage system based on water quality and quantity

Abstract

The application relates to the field of urban sewage system quality improvement and efficiency evaluation and intelligent water management operation technology, and discloses a method for evaluating the quality improvement and efficiency effect of a sewage system based on water quality and quantity. The method comprises the following steps: acquiring a drainage pipe network topology and a monitoring point distribution, establishing a corresponding relationship between analysis partitions and monitoring points; constructing an upstream and downstream topology relationship model between monitoring points; according to measured flow and pollutant concentration data, analyzing the actual total pollutant emission of each analysis partition along the topology relationship from the periphery to the source; multiplying the preset target concentration and the calculated flow to obtain the target total emission; comparing the actual and target total emissions to obtain the pollutant load gap value, and evaluating the quality improvement and efficiency effect of each analysis partition according to the pollutant load gap value. The application uses the topology relationship to analyze the measured data step by step, can separate the actual pollutant emission of each partition from the mixed monitoring data, and realizes independent quantitative evaluation of the collection efficiency of the partition.

Description

Technical Field

[0001] This application relates to the fields of urban wastewater system quality improvement and efficiency enhancement assessment and smart water management operation and maintenance technology, and in particular to a method for assessing the effectiveness of wastewater system quality improvement and efficiency enhancement based on water quality and quantity. Background Technology

[0002] With the deepening of efforts to improve the quality and efficiency of urban wastewater treatment, accurately assessing the actual total amount of pollutants discharged and the load gap within drainage zones has become a key basis for determining the operational health of pipe networks and identifying problems such as wastewater infiltration or rainwater infiltration. However, urban drainage pipe network systems are typical complex dynamic systems, and their operational status is affected by a combination of external factors, including rainfall runoff, downstream water level backwater, groundwater level fluctuations, and intermittent industrial wastewater discharge. In actual engineering projects, the water quantity and quality transport processes within the pipe network often exhibit significant non-steady-state characteristics. This high degree of dynamic change makes it difficult for data based on a single monitoring section to truly reflect the overall pollutant discharge situation of the entire zone, posing a significant challenge to accurate load calculation.

[0003] Currently, the industry commonly employs either a back-calculation method based on the principle of mass conservation or a statistical analysis method based on monthly averages to assess pollutant loads in drainage zones. The former sets a uniform theoretical pollutant concentration based on the pipe network topology and uses data from the total outlet to back-calculate the theoretical total pollutant load for each zone upstream. The latter estimates pollutant flux by multiplying the long-term average flow rate by the average concentration. These methods, under ideal conditions where the pipe network operates relatively stably and boundary conditions are clear, can provide a certain reference and are currently the basic means of regional pollution source apportionment.

[0004] However, the above methods have the following limitations in practical applications: First, traditional calculation methods are generally based on the assumption of steady-state operation of the pipeline network, ignoring the surge in flow and dilution effect of pollutant concentration caused by rainwater infiltration during rainfall. They also cannot accurately simulate abnormal fluctuations in pollutant concentration caused by the scouring of sediments within the pipeline during high water levels, leading to significant deviations between the calculation results and actual discharge conditions. Second, existing models struggle to effectively identify losses or abnormal gains of pollutants during transport, such as sewage seepage or illegal overflows caused by pipeline damage. The simple upstream-downstream difference calculation logic will classify these physical flow errors as calculation errors, resulting in distorted assessment results. Finally, traditional methods lack effective fusion of multi-source heterogeneous data, failing to isolate abnormal operating conditions from complex monitoring time-series data, making it difficult to accurately identify the true causes of load gaps and meet the needs of refined governance. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, this application provides a method for evaluating the quality improvement and efficiency enhancement effect of a wastewater system based on water quality and quantity, overcoming the following problems existing in the prior art:

[0007] Existing methods generally rely on the assumption of steady-state operation of the pipeline network to make overall back-inferences, ignoring the surge in flow and dilution of pollutant concentration caused by rainwater infiltration during rainfall, as well as the abnormal fluctuations in pollutant concentration caused by pipeline sediment scouring during high water levels, resulting in significant deviations between the assessment results and the actual discharge situation.

[0008] Existing methods are based on simple upstream and downstream difference calculation logic, which makes it difficult to distinguish between physical losses such as sewage seepage and illegal overflow caused by pipeline damage and calculation errors, resulting in distorted assessment conclusions and a lack of reliability in guiding subsequent treatment measures.

[0009] Existing methods rely solely on monthly average values ​​of flow and concentration from a single dimension for statistical analysis. This makes it impossible to identify and isolate interference from abnormal operating conditions from complex monitoring time-series data, making it difficult to accurately determine the true causes of pollutant load gaps and failing to meet the needs of refined investigation and long-term operation and maintenance.

[0010] (II) Technical Solution

[0011] To achieve the above objectives, the main technical solutions adopted in this application include:

[0012] This application provides a method for evaluating the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity, the specific steps of which include:

[0013] S1. Obtain multi-source data on the sewage collection service area covered by a sewage treatment plant and its supporting pipe network, including the topology of the drainage pipe network, flow direction data and distribution of monitoring points;

[0014] Based on the drainage network topology, monitoring point distribution, and the network confluence boundary and pump station service area, each analysis zone is divided level by level, and the correspondence between each analysis zone and each monitoring point is established; wherein, the network confluence boundary and pump station service area are determined analytically based on the drainage network topology.

[0015] S2. Based on the drainage network topology and flow direction data, construct an upstream and downstream topological relationship model between the monitoring points; obtain the measured flow data of each monitoring point, and along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point, analyze the measured flow of each monitoring point into the calculated sewage flow of each analysis zone step by step.

[0016] S3. Obtain pollutant concentration monitoring data at each monitoring point, combine the calculated wastewater flow rate of each analysis zone with the upstream and downstream topological relationship model, and analyze and calculate the actual total pollutant discharge of each analysis zone step by step along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point.

[0017] S4. Multiply a preset target concentration of pollutants by the calculated wastewater flow rate of each analysis zone to obtain the total target discharge of pollutants for each analysis zone.

[0018] S5. Compare the actual total pollutant emissions of each analysis zone with the target total pollutant emissions, calculate the pollutant load gap value, and determine the problem zones with insufficient pollutant collection and their investigation priority ranking in each analysis zone based on the pollutant load gap value.

[0019] Optionally, in some embodiments of this application, the wastewater collection service area is based on the wastewater treatment plant as the total catchment node and assessment benchmark.

[0020] The analysis zones are drainage sub-regions divided upstream of the wastewater treatment plant based on the pipe network confluence boundary and the service range of the pumping station; the boundaries between each analysis zone are consistent with the pipe network confluence path, and the wastewater in each analysis zone flows out through the corresponding monitoring point;

[0021] The monitoring point for each analysis zone is located in the inspection well before the main pipe of that analysis zone connects to the next level of the pipeline network; the water quantity and water quality data measured at each monitoring point are the total water quantity and total water quality after mixing the water from that monitoring point and all upstream analysis zones.

[0022] Optionally, in some embodiments of this application, the step of S2, when acquiring the measured flow data of each monitoring point, further includes: acquiring the rainfall data of the sewage collection service area within a preset assessment period, and dividing the time period within the preset assessment period into dry period and rainy period based on the rainfall data;

[0023] In S3, when analyzing and calculating the actual total pollutant emissions of each analysis zone step by step, the measured flow data and pollutant concentration monitoring data corresponding to the dry weather period and the rainy weather period are used respectively to calculate the actual total pollutant emissions of each analysis zone under dry weather conditions and rainy weather conditions.

[0024] In step S5, when determining the problem zones with insufficient pollutant collection and their investigation priorities in each analysis zone based on the pollutant load gap value, the actual total amount of pollutant emissions under dry and rainy conditions is compared to quantify the impact of rainfall on the pollutant collection efficiency of each analysis zone.

[0025] Optionally, in some embodiments of this application, the step S2 of constructing the upstream and downstream topological relationship model between the monitoring points further includes:

[0026] Obtain the physical property data of each pipe section, including pipe diameter, pipe length, pipe material roughness, and effective flow cross-sectional area calculated based on pipe diameter;

[0027] The physical attribute data is embedded as a weighting factor into the topological relationship model to construct a weighted network hydraulic model;

[0028] Acquire historical or real-time pipeline endoscopic image data corresponding to each analysis partition;

[0029] The pre-trained convolutional neural network model is used to identify the endoscopic images of the pipe to determine the siltation coverage and structural defect coefficient inside the pipe.

[0030] Based on the siltation coverage rate and structural defect coefficient, the effective flow cross-sectional area and pipe roughness in the physical attribute data are dynamically corrected, and the corrected physical attribute data is embedded in the weighted pipe network hydraulic model for weight allocation correction of the measured flow data at each monitoring point.

[0031] Optionally, in some embodiments of this application, the step-by-step calculation of the wastewater flow rate of each analysis zone in step S2 follows the following flow balance relationship:

[0032] ;

[0033] in, For analyzing partition S j Calculate the wastewater flow rate, Q j Let D(j) be the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, and Q be the measured flow rate of node j. k This represents the measured flow rate at upstream monitoring node k.

[0034] Optionally, in some embodiments of this application, the step-by-step analysis and calculation of the actual total pollutant emissions for each analysis zone in step S3 follows the following pollutant mass balance relationship for any monitoring node j in the pipeline network:

[0035] ;

[0036] Among them, M j For analyzing partition S j The actual total amount of pollutants emitted, C j For monitoring pollutant concentration data at monitoring node j, Q jTo monitor the measured flow of node j, D(j) is the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, C k Q represents the pollutant concentration monitoring data at upstream monitoring node k. k This represents the measured flow rate at upstream monitoring node k.

[0037] Optionally, in some embodiments of this application, step S5, based on the pollutant load deficit value, determines the problem zones with insufficient pollutant collection in each of the analysis zones and their investigation priority ranking, including:

[0038] Calculate the following density indices for each analysis partition, and prioritize the issues in each analysis partition based on these density indices:

[0039] Load gap per unit area MA j The calculation formula is: MA j = M j / A j ;

[0040] Load gap per unit pipeline length ML j The calculation formula is: ML j = M j / L j ;

[0041] Full-pipe unit operating water depth load gap M Hj The calculation formula is: M Hj = M j / H j ;

[0042] in, M j For analyzing partition S j The pollutant load deficit value, A j For analyzing partition S j area, L j For analyzing partition S j The length of the sewage pipe network within the area, H j For analyzing partition S j The average operating water depth in a fully operational sewage pipe network;

[0043] The higher the value of the density index, the higher the priority of the corresponding analysis and screening.

[0044] Optionally, in some embodiments of this application, after S5, the following steps are also included:

[0045] Obtain water quality characteristic factor data for each monitoring point, wherein the water quality characteristic factor includes at least one of ammonia nitrogen concentration, total phosphorus concentration, and conductivity.

[0046] Based on the water quality characteristic factor data and the actual total amount of pollutants discharged in each analysis zone, the composition of pollution sources in each analysis zone is analyzed to identify at least one abnormal situation, such as leakage of domestic sewage, intrusion of industrial wastewater, or infiltration of groundwater.

[0047] When the actual total pollutant discharge of the analysis zone is lower than the target total pollutant discharge of the analysis zone, and the conductivity is lower than the preset groundwater characteristic threshold, it is determined that there is groundwater infiltration in the analysis zone.

[0048] When the actual total pollutant discharge of the analysis zone is higher than the target total pollutant discharge of the analysis zone, and the combination of water quality characteristic factors is abnormal, it is determined that there is industrial wastewater intrusion in the analysis zone.

[0049] Optionally, in some embodiments of this application, after S5, the following step is further included:

[0050] Based on the pre-constructed flow conservation equation and pollutant mass conservation equation under abnormal operating conditions, combined with the pollutant load deficit value and the water quality characteristic factor data, the main causes of abnormal operating conditions in each analysis zone are analyzed, and corresponding treatment measures are selected from the preset measure library accordingly.

[0051] Optionally, in some embodiments of this application, the flow conservation equation is:

[0052] ;

[0053] The mass conservation equation for the pollutants is:

[0054] ;

[0055] Where Q1 and C1 are the measured wastewater volume and measured pollutant concentration at the end of the drainage section under abnormal operating conditions, respectively; Q0 ​​and C0 are the measured wastewater flow rate and measured pollutant concentration at the upstream input section, respectively; Q 清 C 清 These represent the volume and concentration of clean water flowing into the sewage system under abnormal operating conditions caused by mixed rainwater and sewage connections; Q 地 C 地 These represent the amount of groundwater infiltrating at the location of pipeline rupture under abnormal operating conditions and the background concentration of groundwater; Q 调 C 调These represent the volume of overflow wastewater collected and returned to the wastewater system by the storage facilities and its average pollutant concentration; Q 渗i C 渗i These represent the volume of sewage leaking from the i-th pipe rupture under abnormal operating conditions and the sewage concentration at the leak point; Q 溢j C 溢j These represent the overflow sewage volume and sewage concentration at the j-th location under abnormal operating conditions caused by combined sewer overflow or pipeline overload; W 沉 This refers to the amount of insoluble pollutants deposited in the pipes during high-water-level operation of the pipeline network under abnormal operating conditions; W 消 This represents the amount of soluble pollutants lost through biodegradation during pipeline transportation; n represents the total number of sewage seepage points in the pipeline network, and m represents the total number of overflow points in the pipeline network.

[0056] (III) Beneficial Effects

[0057] The method for assessing the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity, as proposed in this application, can achieve the following beneficial effects:

[0058] By establishing a one-to-one correspondence between analysis zones and monitoring points, and combining upstream and downstream topological relationship models for step-by-step analytical calculations, the actual total pollutant emissions of each analysis zone can be extracted from mixed monitoring data, solving the problem that existing methods cannot accurately calculate the emissions of each independent zone. By directly using measured flow data and pollutant concentration monitoring data from each monitoring point as input, and analyzing step-by-step from the end to the source along the pipeline topology, the calculation bias caused by relying on theoretical sewage concentration, clean water concentration, and other assumed parameters in existing methods is reduced, making the assessment results closer to the actual operating state of the drainage system. By comparing the actual total pollutant emissions of each analysis zone with the target total pollutant emissions, the pollutant load gap value is calculated, which can intuitively reflect the actual contribution of each analysis zone to the overall collection efficiency, facilitating the determination of the priority of problem zones and providing a basis for the precise deployment of geophysical exploration and remediation projects.

[0059] All the data required for this application can be obtained from the smart platform. The calculation process is automatically completed based on the topological relationship model, which reduces the reliance on manual topology sorting and manual calculation. It is suitable for deployment and operation in the smart water system and supports regular evaluation and dynamic tracking in the long-term operation and maintenance phase. Attached Figure Description

[0060] Figure 1 This is a flowchart of a method for assessing the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity, as described in this application.

[0061] Figure 2 This is a schematic diagram of the drainage network topology and analysis partitions in one embodiment of this application;

[0062] Figure 3 This is a schematic diagram of material migration in a pipeline network under normal operating conditions in one embodiment of this application;

[0063] Figure 4 This is a schematic diagram of material migration under abnormal operating conditions in a drainage network, as shown in one embodiment of this application. Detailed Implementation

[0064] To better explain and facilitate understanding of this application, a detailed description of the application is provided below with reference to the accompanying drawings and specific embodiments. In the prior art, the evaluation methods for improving the quality and efficiency of wastewater systems can be mainly summarized into the following three categories:

[0065] The first category is the overall back-calculation estimation method: for example, CN119041537A. This method collects data on regional sewage facility influent, pump station operation, and water consumption, and uses water balance and mass conservation formulas to perform overall back-calculation of the entire region to estimate the amount of external water infiltration and the concentration of water quality after renovation. This type of method relies on assumed parameters such as theoretical sewage volume and original sewage concentration. Deviations in parameter values ​​directly affect the reliability of the calculation results; the definitions of some terms in the formula may differ in different application scenarios; and since it does not perform calculations for each independent zone, it cannot accurately locate the source of the problem, thus limiting its role in guiding troubleshooting.

[0066] The second type is the theoretical concentration substitution method: For example, CN114240127A, based on the measured water quality and quantity data at the end of each drainage zone, artificially sets uniform theoretical sewage concentration and theoretical clean water concentration, substitutes them into the mass conservation formula to deduce the theoretical sewage volume and theoretical clean water volume of each zone in one go, and then evaluates the effect of the quality improvement and efficiency enhancement project. This type of method relies on artificially set theoretical concentration values, which vary greatly in different regions, raising questions about the objectivity of the results; the calculation only performs a one-time overall deduction, without separating the data step by step along the pipeline topology; the evaluation conclusion is based on the level of water quality concentration, rather than the total amount of pollutants discharged, and is disconnected from the actual assessment indicators.

[0067] The third category is conventional intelligent operation and maintenance methods, which rely on intelligent water management platforms to visualize and statistically analyze monitoring data, identifying problem areas by comparing water quality concentrations at various monitoring points. This type of method focuses only on water quality concentration, ignoring the contribution of water volume to the total amount of pollutants collected; it only evaluates each monitoring point independently, lacking upstream and downstream correlation analysis based on topological relationships; and the output results are vague conclusions such as qualitatively low concentrations, failing to provide specific pollutant load deficit values ​​for each zone and a priority ranking for investigation.

[0068] To address this, this application provides a method for evaluating the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity. By establishing a correspondence between analysis zones and monitoring points, constructing an upstream and downstream topological relationship model between monitoring points, and analyzing the calculated wastewater flow and actual total pollutant discharge of each zone step by step from the end to the source along the topological relationship, the method uses the total pollutant discharge rather than concentration as the evaluation index, thereby achieving independent quantitative evaluation of the collection efficiency of each analysis zone and prioritizing the issues, overcoming the shortcomings of the aforementioned prior art.

[0069] To better explain and facilitate understanding of this application, a detailed description of its embodiments is provided below in conjunction with the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a clearer and more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0070] Example 1:

[0071] Figure 1 According to this application, a method for assessing the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity is proposed, such as... Figure 1 As shown, the method specifically includes:

[0072] S1. Obtain multi-source data on the sewage collection service area covered by a sewage treatment plant and its supporting pipe network, including the topology of the drainage pipe network, flow direction data and distribution of monitoring points;

[0073] Based on the drainage network topology, monitoring point distribution, and the network confluence boundary and pump station service area, each analysis zone is divided level by level, and a correspondence between each analysis zone and each monitoring point is established; wherein, the network confluence boundary and pump station service area are determined analytically based on the drainage network topology; the monitoring point corresponding to each analysis zone is located in the inspection well before the main pipe of that analysis zone connects to the next level of the network;

[0074] S2. Based on the drainage network topology and flow direction data, construct an upstream and downstream topological relationship model between the monitoring points; obtain the measured flow data of each monitoring point, and along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point, analyze the measured flow of each monitoring point into the calculated sewage flow of each analysis zone step by step.

[0075] S3. Obtain pollutant concentration monitoring data at each monitoring point, combine the calculated wastewater flow rate of each analysis zone with the upstream and downstream topological relationship model, and analyze and calculate the actual total pollutant discharge of each analysis zone step by step along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point.

[0076] S4. Multiply a preset target concentration of pollutants by the calculated wastewater flow rate of each analysis zone to obtain the total target discharge of pollutants for each analysis zone.

[0077] S5. Compare the actual total pollutant emissions of each analysis zone with the target total pollutant emissions, calculate the pollutant load gap value, and determine the problem zones with insufficient pollutant collection and their investigation priority ranking in each analysis zone based on the pollutant load gap value.

[0078] This embodiment establishes a correspondence between analysis zones and monitoring points, and analyzes the data step by step from the source to the end along the upstream and downstream topological relationship model. This allows for the extraction of the actual total pollutant emissions of each analysis zone from mixed monitoring data, overcoming the shortcomings of existing technologies that can only estimate the overall system and cannot accurately locate problem zones. It directly uses measured flow data and pollutant concentration monitoring data from each monitoring point as input, reducing calculation biases caused by assumptions such as theoretical wastewater concentration and clean water concentration in existing technologies. Simultaneously, the step-by-step analysis process follows the actual pipe network topology, making the evaluation results closer to the actual operating state of the drainage system. By comparing the actual total pollutant emissions of each analysis zone with the target total emissions, rather than just comparing pollutant concentrations, it avoids omissions due to insufficient water volume despite meeting concentration standards. The evaluation conclusions are directly linked to the engineering requirements for total pollutant collection, providing more accurate guidance for subsequent treatment measures. By calculating the pollutant load gap value of each analysis zone, it can intuitively reflect the actual contribution of each zone to the overall collection efficiency, facilitating the determination of the priority of problem zones and providing a basis for precise deployment of geophysical detection and remediation projects.

[0079] Example 2:

[0080] This embodiment takes the sewage collection service area covered by the Chengdong Sewage Treatment Plant and its supporting pipe network in a certain city as the overall evaluation object. The sewage treatment plant is responsible for treating domestic and some industrial wastewater in the Chengdong area, and the total length of the existing drainage pipe network in the area is about 85km.

[0081] Before the assessment began, the Chengdong area had already conducted a comprehensive survey of the drainage pipe network and constructed a GIS system, inputting physical attribute data such as pipe connection relationships, pipe diameter, pipe material, and pipe bottom elevation into the smart platform. The smart platform communicates with online flow meters, online water quality analyzers, level gauges, and other equipment at each monitoring point through data interfaces to achieve real-time data entry. Asset data such as pipeline CCTV endoscopic inspection images and historical maintenance records are uniformly archived in the smart platform.

[0082] This embodiment relies on the intelligent platform to divide the service area of ​​the Chengdong Wastewater Treatment Plant into eight independent analysis zones (numbered S0 to S7) according to the pipe network confluence boundary and the pump station service area. Among them, S1 to S7 are drainage sub-zones divided step by step according to the pipe network topology, and S0 is the terminal water collection area before entering the wastewater treatment plant (W0) after the confluence of each zone.

[0083] Each analysis zone has a unique corresponding monitoring method. For analysis zones S1 to S7, water quality and quantity monitoring points (numbered J1 to J7) are set up in key inspection wells before their main pipes connect to the next-level pipeline network. For analysis zone S0, the online monitoring data from the wastewater treatment plant inlet (W0) is used directly as its corresponding monitoring basis. The layout of monitoring points follows the "zone outlet" principle. The water volume and pollutant concentration measured at this monitoring point (or inlet) reflect the total water volume and total water quality of the analysis zone and all upstream zones after mixing.

[0084] The entire calculation process is completed using the smart water management platform. At the beginning of each month, the platform triggers an evaluation command. The following is a detailed explanation of this embodiment, with specific technical steps.

[0085] 1. Acquisition of basic data and partitioning;

[0086] 1.1. Obtain basic data from the intelligent platform;

[0087] At the start of the assessment, technicians logged into the smart water management platform, which retrieved and exported the following multi-source data from the Chengdong area from a unified database:

[0088] (1) Drainage pipe network GIS data: includes physical attribute information such as the starting / ending manhole number, pipe diameter, pipe material, pipe bottom elevation, pipe length, and roughness coefficient of each pipe section. The smart platform extracts the pipe network topology and sewage flow direction data from it.

[0089] (2) Monitoring point distribution data: including the installation well numbers, equipment types, and geographical coordinates of the online flow meters (including liquid level measurement modules) and online water quality analyzers;

[0090] (3) Pipeline asset data: including the year of laying of each pipe section, historical maintenance records, and archived CCTV endoscopic inspection images;

[0091] (4) Meteorological and hydrological data: including daily rainfall and hourly rainfall intensity data of the target area in the past and current assessment period, as well as liquid level monitoring data of each monitoring point.

[0092] Of the aforementioned data, dynamic monitoring data is transmitted to the database in real-time or periodically via data interface communication between the intelligent platform and the data transmission units of each monitoring device; static data (such as GIS data and asset data) is pre-entered and updated through the platform's data maintenance module. After data extraction, the platform automatically performs data cleaning and format preprocessing to prepare for subsequent calculations.

[0093] 1.2. Define the analysis partitions;

[0094] The total length of the pipeline network in the eastern part of the city is approximately 85 kilometers. Based on the pipeline connection relationships in the drainage network GIS data, the intelligent platform traces along the pipeline route and automatically analyzes and determines the upstream water catchment area of ​​each confluence node, forming the pipeline confluence boundary; at the same time, based on the location of the pumping station and the coverage area of ​​all upstream confluence pipelines, it comprehensively determines the service area of ​​the pumping station.

[0095] Based on the aforementioned pipe network confluence boundaries and pump station service areas, the Chengdong area is divided into eight independent analysis zones (numbered S0 to S7). Among them, S1 to S7 are drainage sub-zones, and S0 is the terminal catchment area before the W0 inlet of the sewage treatment plant.

[0096] Figure 2 This diagram illustrates the drainage network topology and analysis zones in this embodiment. The boundaries of each zone are consistent with the actual confluence paths of the network, ensuring that all sewage within each zone flows out through its corresponding outlet, and there is no direct confluence across zones. Except for S0, each analysis zone (S1-S7) has corresponding monitoring points (J1-J7) in the key inspection wells before the main pipe connects to the next-level network. The monitoring data for S0 is based on the data from the sewage treatment plant inlet (W0).

[0097] 1.3 Establish the correspondence between analysis zones and monitoring points;

[0098] Each analysis zone has a monitoring point at the end of its main pipeline, numbered J1 to J7, corresponding one-to-one with analysis zones S1 to S7. Analysis zone S0 corresponds to the monitoring data of the wastewater treatment plant's W0 inlet. Among them, J1 and J2 are the primary confluence monitoring nodes for the area, and J3 to J7 are secondary and terminal monitoring nodes.

[0099] Monitoring points J1 to J7 are all located in key inspection wells before the main pipeline of the corresponding zone merges into the next-level pipeline network. This location ensures that the water quantity and quality data measured at each monitoring point accurately reflect the total water quantity and quality after mixing with the water from the node and all upstream analysis zones.

[0100] 2. Construct a topological relationship model of upstream and downstream monitoring points;

[0101] 2.1. Methods for constructing topological relationships;

[0102] based on Figure 2 The pipeline flow direction hierarchy shown is used to construct the upstream and downstream topological relationships between monitoring nodes. The specific method is as follows: virtual connecting pipelines are introduced between each monitoring point (J1~J7), and virtual intermediate nodes are added at the intersections of pipelines at non-monitoring points; by traversing all connecting pipelines and pipeline flow directions, the strict upstream and downstream hierarchy of each monitoring node is determined.

[0103] Taking this embodiment as an example: terminal nodes: J3, J4, J5, and J7 are terminal monitoring nodes, and there are no other monitoring points directly upstream of them;

[0104] Intermediate levels: J3, J4, and J5 converge to J1; J7 converges to J6, and J6 then converges to J2;

[0105] Final confluence: J1 and J2 serve as primary confluence nodes, and their effluent ultimately flows into wastewater treatment plant W0 (corresponding to the data collection point of monitoring point J0).

[0106] 2.2. Construct a weighted hydraulic model of the pipeline network;

[0107] First, the physical property data of each pipe segment are obtained, including pipe diameter, pipe length, pipe material roughness, and the effective flow cross-sectional area calculated based on the pipe diameter. This physical property data is then embedded as weighting factors into the topology model to construct an initial weighted network hydraulic model.

[0108] Furthermore, to correct parameter deviations caused by pipe aging or siltation, historical pipe endoscopic image data corresponding to each analysis partition in the intelligent platform are retrieved. A pre-trained convolutional neural network model is used for automated identification to determine the siltation coverage rate and structural defect coefficient within the pipe. The pre-trained convolutional neural network model is trained offline using a large number of labeled pipe endoscopic sample images. These samples include typical pipe images with different diameters, varying degrees of siltation, and various structural defects. The model convergence training is completed through feature extraction and sample calibration. After training, the model is permanently deployed to the intelligent platform, enabling automated identification and detection of newly added endoscopic images.

[0109] Based on the aforementioned siltation coverage rate and structural defect coefficient, the initial hydraulic model is dynamically modified: the effective flow cross-sectional area of ​​the pipe section is reduced according to the siltation coverage rate; the equivalent roughness of the pipe section is dynamically adjusted according to the structural defect coefficient (such as rupture, corrosion, etc.) to reflect the impact of changes in pipe wall condition on flow resistance. The modified physical property data is then re-embedded into the weighted network hydraulic model for high-precision weight allocation and analysis of the measured flow data at each monitoring point.

[0110] Specifically, this embodiment utilizes a pre-trained convolutional neural network (CNN) model within the intelligent platform to automatically identify pipe endoscopic images in sections S0 and S1. The model output shows that the average siltation coverage in section S0 reaches 30%, with multiple rupture defects (structural defect coefficient of 0.8). Based on this coefficient, the effective flow cross-sectional area of ​​the pipe section is dynamically corrected, resulting in a 10% improvement in flow rate resolution accuracy.

[0111] The pre-trained convolutional neural network model is trained offline using historical pipe endoscope image samples (covering different pipe diameters, different degrees of siltation, and various structural defects). The model converges through feature extraction and sample labeling. After training, it is deployed to the intelligent platform for automated recognition of newly acquired endoscope images.

[0112] 2.3. Topological relationship verification;

[0113] Real-time liquid level data from each monitoring point is synchronously acquired from the intelligent platform, and the difference in liquid level elevation between adjacent monitoring points is calculated to reflect the actual hydraulic gradient of the pipeline network. The flow direction determined based on the actual hydraulic gradient is compared and verified with the theoretical pipeline flow direction marked in the GIS system: if the two determine the flow direction is consistent, the existing topology model remains unchanged; if the two determine the flow direction is inconsistent, a topology anomaly alarm is automatically generated, and the upstream and downstream topology relationship model is dynamically corrected according to the actual hydraulic gradient direction. Through the above dynamic verification and correction mechanism, it is ensured that the topology model truly reflects the pipeline network operating status, thereby guaranteeing the accuracy of subsequent flow analysis and pollutant recursive calculation.

[0114] 3. Monitoring data preprocessing and division of drought and rainy seasons;

[0115] 3.1 Obtain monitoring data;

[0116] The assessment period was selected as July of a certain year, which coincides with the region's main flood season, characterized by frequent short-term heavy rainfall and continuous rainy weather, easily leading to complex conditions such as combined sewer overflows, rainwater backflow, high water levels in the pipe network, and pollutant deposition, scouring, and resuspension. At the beginning of July, the monthly average flow rate and monthly average BOD5 (five-day biochemical oxygen demand) concentration of each monitoring point within this assessment period were extracted from the smart water management platform in advance, along with rainfall data for the same period.

[0117] In this embodiment, BOD5 is uniformly used as the pollutant accounting indicator. Its selection is based on the following criteria: the centralized collection rate assessment of urban domestic sewage uses BOD5 as the measurement benchmark. Using BOD5 accounting can directly assess whether the water quality and quantity entering the sewage treatment plant from each analysis zone meet the collection rate requirements. If only COD data can be obtained under on-site monitoring conditions, the characteristic B / C ratio of each monitoring point or concentration range is determined through multiple preliminary sampling. Based on this, the COD monitoring value is converted into a BOD5 value before being used in the calculation.

[0118] By acquiring flood season monitoring data in batches at the beginning of the month and completing the assessment, we can identify the shortcomings in pollutant collection and high-risk areas in each analysis zone in advance. This will guide the pipeline inspection, CCTV spot checks, pump station scheduling and operation and maintenance management during the flood season in July, enabling advance prediction and precise control of improving the quality and efficiency of the sewage system during the flood season.

[0119] It should be noted that in this embodiment, all monitoring data is uniformly acquired from online monitoring equipment or manual sampling records at each monitoring point through the intelligent platform. The data acquisition frequency, transmission protocol, and storage format of each monitoring point are uniformly managed by the intelligent platform, and the platform automatically completes data extraction, time-stamp alignment, and outlier marking during evaluation.

[0120] 3.2. Process the acquired data and simultaneously execute the following quality control steps;

[0121] (1) Time-series anomaly and missing data detection and repair: detect abnormal jumps and missing values ​​in the time series of data at the same monitoring point, and verify the sampling time synchronization of data between different monitoring points; for the detected abnormal or missing data, based on the upstream and downstream topology relationship model, and using the principle of mass conservation, hydraulic simulation is performed using the effective data of adjacent monitoring points (upstream inflow point or downstream receiving point) in the same time period to presumably repair the abnormal or missing data, thereby restoring the real pipeline network operation status to the greatest extent.

[0122] (2) Rainfall impact assessment, dividing dry / rainy days into different time periods;

[0123] Rainfall data within the wastewater collection service area is obtained within a preset assessment period, and the time period within the assessment period is divided into dry and rainy periods. In this embodiment, a rainy period is defined as a daily rainfall of ≥5mm, and a dry period is defined as a daily rainfall of <5mm.

[0124] During rainy periods, the mixing of rainwater runoff dilutes wastewater concentration and easily causes pipe network overflows, leading to the loss of pollutants midway. Directly deducting the increase in rainwater runoff would result in significant errors. Therefore, this embodiment does not forcibly deduct or correct the measured data for rainy days. Instead, it retains two sets of measured flow and concentration data for dry and rainy days, respectively, keeping the rainy day data as an independent sample for subsequent calculation of the proportion of total pollutant changes in each zone under different rainfall intensities.

[0125] By comparing the ratio of total pollutants on rainy days to those on dry days, the degree of combined sewer overflow and the risk of overflow in each analysis zone are ranked and assessed. The larger the deviation in the ratio, the more severe the impact of combined sewer overflow or overflow.

[0126] Based on experience, this embodiment sets the following judgment rules: when the pollutant load gap of the analysis zone increases by more than 30% after rainfall compared to before rainfall, it is judged as a rainwater mixing anomaly; if the load gap is stable for a long time without significant fluctuations, it is judged as a continuous groundwater infiltration anomaly.

[0127] 4. Wastewater flow rate is calculated and analyzed step by step for each analysis zone;

[0128] 4.1. Calculate the wastewater flow rate for each zone step by step using analytical methods;

[0129] Based on the principle of pipeline flow continuity (i.e., the law of conservation of mass), a strategy of deducting flow from the end to the source is adopted to analyze the actual contribution flow of each analysis zone. The specific calculation logic is as follows:

[0130] For terminal monitoring nodes, there are no other monitoring nodes upstream, and the measured flow rate is the calculated sewage flow rate for the corresponding analysis zone. For non-terminal monitoring nodes, the measured flow rate includes upstream water, and the calculated sewage flow rate for the corresponding analysis zone is equal to the measured flow rate of the node minus the sum of the flows of all its directly upstream monitoring nodes.

[0131] The calculation formula is: ;

[0132] in, For analyzing partition S j Calculate the wastewater flow rate, Q j Let D(j) be the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, and Q be the measured flow rate of node j. k This represents the measured flow rate at upstream monitoring node k.

[0133] 4.2. Substitute engineering values ​​for calculation and verification:

[0134] (1) Terminal monitoring node: There are no other upstream monitoring nodes, and its zoned calculated flow is equal to the node's measured flow:

[0135] QS3=Q3=1200m 3 / d;QS4=Q4=1200m 3 / d;

[0136] QS5=Q5=1200m 3 / d;QS7=Q7=1600m 3 / d;

[0137] (2) Intermediate confluence monitoring node: The independent calculated flow rate of this partition is obtained by subtracting the upstream terminal node flow rate from the downstream measured flow rate according to the topology.

[0138] QS6 = Q6 - Q7 = 1600m 3 / d;QS1=Q1-(Q3+Q4+Q5)=1300m 3 / d;

[0139] QS2 = Q2 - Q6 = 1700m 3 / d;

[0140] (3) Main busbar node in front of the plant:

[0141] QS0 = Q0 - (Q1 + Q2) = 0m 3 / d;

[0142] Where QS0=0m 3 / d indicates that the main confluence node in front of the plant (i.e., the inlet of the wastewater treatment plant) has no additional independent flow generation zone. The wastewater in the area is formed by the confluence of the upstream analysis zones, and the flow topology is closed and balanced.

[0143] 5. Step-by-step analysis of the actual total emissions of pollutants in each analysis zone;

[0144] 5.1. Calculate the actual total amount of pollutants emitted;

[0145] Following the upstream-to-downstream step-by-step deduction logic consistent with flow analysis, the actual total pollutant emissions for each zone are calculated based on pollutant mass balance. The calculation formula is as follows:

[0146] ;

[0147] Among them, M j For analyzing partition S j The actual total amount of pollutants emitted (usually in kg / d), C j Q represents the BOD5 pollutant concentration monitoring data (mg / L) at monitoring node j. j To monitor the measured flow rate (m) of node j 3 / d), D(j) is the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, C k Q represents the pollutant concentration monitoring data (mg / L) at upstream monitoring node k. k The measured flow rate (m³) at upstream monitoring node k 3 / d).

[0148] Specifically, in this embodiment, BOD5 is selected as the characteristic pollutant indicator. The average concentration measured within the assessment period is used for each monitoring node. Taking zone S1 as an example, the calculation process of its actual total pollutant emissions is as follows: The average BOD5 concentration measured at monitoring node J1 is 180 mg / L. Combined with the zone's calculated flow rate QS1 = 1300 m³ / L obtained above, the total flow rate is calculated. 3 Substituting / d into the pollutant mass balance formula, the actual total emission M of the original pollutants in zone S1 without model correction can be calculated.S1 .

[0149] The same calculation logic was used for the remaining analysis partitions. Based on the measured concentrations of the corresponding nodes and the calculated flow rates of the partitions, the actual total emissions of the original pollutants in each partition were obtained one by one.

[0150] 6. Calculation of total target emissions of pollutants for each analysis zone;

[0151] Based on local wastewater quality improvement and efficiency enhancement assessment standards, a unified assessment target BOD5 concentration is set: C T =100mg / L. The target emission concentration is calculated by multiplying the flow rate by the target concentration for each zone to obtain the total target emission of pollutants for each zone: MT j =C T ×QS j .

[0152] Among them, MT j Indicates analysis partition S j The target total emissions of pollutants (kg / d);

[0153] C T This indicates the set target BOD5 concentration for evaluation (100 mg / L in this example).

[0154] QS j This represents the analysis partition S obtained from the previous analysis. j Calculated flow rate (m 3 / d).

[0155] 7. Calculation and assessment of pollutant load gap;

[0156] 7.1 First, calculate the pollutant load gap for each zone: M j =MT j -M j ;

[0157] when M j When the value is >0, it is determined that the pollutant collection in this analysis zone is insufficient, which may be due to problems such as external water infiltration (occupying up pipe network space), sewage leakage, or rainwater and sewage mixing leading to concentration dilution; when M j When the value is less than 0, the wastewater collection situation in the analysis zone is considered normal, and the preset collection rate target has been met or exceeded.

[0158] like M j Near zero (i.e., actual emissions are roughly equal to target emissions) indicates that the pipeline network in the analyzed area is operating well and the pollutant collection efficiency meets the standards; if M jA value significantly less than zero (i.e., the actual discharge is significantly higher than the target discharge) indicates that the actual collection effect of the analyzed zone is good. There may be two situations: one is that the discharge concentration of the wastewater users in this area is inherently high, which is a normal regional water quality characteristic; the other is that there may be high concentrations of external water (such as industrial wastewater) mixed in, which needs to be further confirmed in combination with water quality characteristic factors.

[0159] It should be noted that the assessment target for the centralized collection rate of urban domestic sewage is usually set at 70%–80%, not 100%, as this already takes into account normal losses during sewage transportation through the pipeline network (such as natural degradation of pollutants and a small amount of seepage). The total pollutant discharge target calculated in this embodiment has been taken according to the assessment requirements; therefore, the actual collection volume being roughly equal to or slightly higher than the target value is within the normal range. Furthermore, from a physical mechanism perspective, backflow into the pipeline network only leads to sewage overflow and a reduction in the collection rate, not an increase in the actual collection volume. Therefore, a significantly higher collection volume primarily indicates the possibility of high-concentration external water intrusion.

[0160] 7.2 Density index assessment;

[0161] To further focus on key areas of investigation and improve treatment efficiency, this embodiment calculates the unit pollution load intensity of each analysis zone and prioritizes the severity of problems in each zone accordingly. The following three priority-ranking density indices are calculated for each analysis zone:

[0162] Load gap per unit area MA j This reflects the degree of pollutant deficiency within a unit catchment area, and the calculation formula is: MA j = M j / A j ;

[0163] Load gap per unit pipeline length ML j This reflects the pollutant collection loss per unit length of pipeline network, and the calculation formula is: ML j = M j / L j ;

[0164] Full-pipe unit operating water depth load gap M Hj This reflects the risk of pollution loss per unit water depth under high water level operating pressure in the pipeline. The calculation formula is as follows: M Hj = M j / H j .

[0165] in, M j For analyzing partition S j The pollutant load deficit value, A j For analyzing partition S j area, L j For analyzing partition S j The length of the sewage pipe network within the area, H j For analyzing partition S j The average operating water depth in a fully operational sewage pipe network.

[0166] Taking into account the load gap per unit area, the load gap per unit pipeline length, the load gap per unit operating water depth in each zone, and the absolute value of the load gap, and combining engineering experience, a comprehensive assessment is conducted to determine the priority of the investigation.

[0167] Furthermore, this embodiment uses measured data from dry and rainy days to independently calculate the actual total emissions of pollutants in each zone under the two operating conditions; the intensity of rainfall disturbance is quantified by the ratio of total pollutants in rainy / dry days; in this embodiment, the ratio of rainy / dry days in zone S1 is 0.85, indicating that rainfall causes a 15% decrease in collection efficiency, and it is given priority for inclusion in the key investigation list of mixed rainwater and sewage.

[0168] Based on calculations and evaluations, the investigation priorities for each analysis partition are determined as follows:

[0169] First priority (extremely high risk): Area S0. This area has the highest load deficit per unit area and the highest load deficit per unit pipeline length in the entire site, and the pipeline length is only 0.8km. The abnormal signals are highly concentrated, indicating that there are serious structural or functional defects in this short pipeline section, requiring immediate engineering intervention.

[0170] Second priority (high risk): Area S1. This area has a large absolute value of pollutant load deficit and a moderate pipeline length, making it a key problem area.

[0171] Third priority (medium risk): S6 area. The pollutant load deficit in this area is positive, but the load deficit density per unit area and per unit pipe length is relatively mild. It is recommended to pay attention to this area in conjunction with routine inspections.

[0172] Fourth priority (low to medium risk): Area S2. This area has a positive pollutant load deficit, indicating insufficient pollutant collection. However, the load deficit per unit area and per unit pipe length is less than that of S6. It is recommended that this area be included in the next quarter's investigation plan, with priority given to investigating suspected points of combined sewer overflow.

[0173] Fifth priority (low risk): S3 area. The pollutant load deficit in this area is close to zero (24g / d), all water quality characteristic factors are normal, and the pipeline network is in good condition. Regular monitoring is sufficient.

[0174] Normal areas (low risk): S4, S5, and S7 areas. The pollutant load deficit in these areas is negative or minimal, pollutant collection efficiency meets standards, and the pipeline network is operating well.

[0175] 7.3 Identification of water quality characteristic factors and abnormal operating conditions;

[0176] In order to accurately quantify the pollutant collection efficiency of each area, this embodiment constructs a material migration model of the pipeline network, corresponding to two operating states: normal operating conditions and abnormal operating conditions.

[0177] like Figure 3 The diagram shown illustrates the material migration in the pipeline network under normal operating conditions as provided in this embodiment. Under ideal or normal operating conditions, pollutants in the pipeline network mainly originate from upstream access nodes and drainage users along the route. During the transportation process, only a small amount of normal physical loss occurs. The water level in the pipeline network is below the full-pipe operating line, and there is no additional infiltration, mixing, seepage, or overflow.

[0178] However, in actual operation, the pipe network is often affected by factors such as external water intrusion, structural defects, and combined sewer overflows, which significantly alters the material conservation relationship. For example... Figure 4 The diagram shown illustrates the migration of materials in a pipeline network under abnormal operating conditions provided in this embodiment. Typical interference processes include:

[0179] Contamination by clean water: Mixing of rainwater pipes, infiltration of groundwater, or the influx of clean water from storage tanks can increase the total flow in the pipe network and dilute the concentration of pollutants.

[0180] Changes in hydraulic conditions: High water levels or full operation of the pipeline network can cause resuspension or new sedimentation processes of deposits in the pipeline, and may even cause sewage to seep out, overflow into rivers through rainwater outlets, or be diverted through regulating reservoirs.

[0181] Intrusion of external pollutants: Abnormal inflow of non-domestic sewage such as industrial wastewater alters the water quality characteristics of the pipe network.

[0182] Based on the above mechanism, in order to accurately calculate the pollutant load gap, this embodiment constructs a pollutant mass conservation equation under abnormal operating conditions and quantifies the impact of various disturbance terms on the material balance.

[0183] (1) Pollutant mass conservation equation under abnormal operating conditions;

[0184] To quantify the impact of various abnormal operating conditions on the material balance of the pipeline network, the following conservation equation is introduced:

[0185] Pollutant mass conservation equation:

[0186] ;

[0187] The flow conservation equation is:

[0188] ;

[0189] Q1 and C1 are the measured sewage volume and measured pollutant concentration at the end of the drainage zone (i.e., the inlet of the sewage treatment plant) under abnormal operating conditions, respectively.

[0190] Q0 and C0 are the measured wastewater flow rate and measured pollutant concentration at the upstream input section (or the outlet of this zone), respectively;

[0191] Q 清 C 清 These represent the volume and concentration of clean water flowing into the sewage system under abnormal operating conditions caused by mixed rainwater and sewage discharge.

[0192] Q 地 C 地 These represent the amount of groundwater infiltrating at the location of pipeline rupture under abnormal operating conditions and the background concentration of groundwater.

[0193] Q 调 C 调 These refer to the volume of overflow wastewater collected and returned to the wastewater system by the storage facilities and its average pollutant concentration.

[0194] Q 渗i C 渗i These are the sewage volume and sewage concentration at the i-th pipe network rupture point under abnormal operating conditions. The water quality varies at different pipe sections. The rupture point can be located by CCTV, and the leakage volume can be estimated by combining the difference in water volume between upstream and downstream and matching the water quality of the corresponding pipe section.

[0195] Q 溢j C 溢j These are the overflow sewage volume and sewage concentration at the j-th overflow point under abnormal operating conditions caused by mixed rainwater and sewage connections or overloaded pipe network. The water quality varies at different overflow points and can be obtained through online monitoring and event recording at the overflow outlet.

[0196] W 沉 This refers to the amount of insoluble pollutants in the sewage that are deposited in the pipes when the pipe network is running at a high water level under abnormal operating conditions. This part of the sediment is easily washed away and overflowed during the rainy season, forming a pollutant storage and collection effect.

[0197] W 消 The amount of biodegradation loss of soluble pollutants during pipeline transportation can be determined by hydraulic residence time, degradation coefficient, or engineering experience data.

[0198] n represents the total number of sewage infiltration points in the pipe network, and m represents the total number of overflow points in the pipe network;

[0199] The above parameters can all be obtained from existing data on the intelligent platform or through conventional testing methods, making them feasible in engineering. It is particularly important to note that Q in the formula... 溢j This refers to the leakage of sewage caused by pipeline damage. Its physical direction is from inside the pipe to outside the pipe, so it is reflected as a negative loss in the mass conservation equation.

[0200] (2) Diagnostic process for S0 partition (first priority);

[0201] First, water quality characteristic factors were obtained from the monitoring points in zone S0. The test results showed that the conductivity of zone S0 was only 650 μS / cm, significantly lower than the characteristic threshold for domestic sewage (e.g., 800 μS / cm). Combined with the calculations above, the pollutant load deficit was positive. M j =598g / d), and the time-series variation characteristics show that the gap is stable in the long term and has no obvious fluctuations.

[0202] Based on this, it is preliminarily determined that there is continuous infiltration of low-conductivity external water (i.e., groundwater) into the area. Substituting into the above conservation equation for analysis, the main manifestation is Q 地 The (groundwater infiltration flow) term increased significantly, which led to the dilution of pollutant concentrations within the system, resulting in a positive load gap.

[0203] (3) S1 partition (second priority) diagnosis:

[0204] The conductivity of zone S1 is within the normal range during dry weather, but the total amount of pollutants varies significantly between dry and rainy days, exhibiting drastic temporal fluctuations. This embodiment uses measured data from dry and rainy days for independent calculations and introduces the "rainy day / dry day pollutant total ratio" for quantitative assessment.

[0205] Calculations showed that the total pollutant load ratio for rainy days to dry days in zone S1 was 0.85, and the pollutant load deficit after rainfall increased by more than 30% compared to before rainfall. Combined with temporal characteristic analysis, the significant increase in the pollutant load deficit after rainfall indicates a rainwater-mixed anomaly. Substituting the data into the flow conservation equation, it was found that during rainfall, Q... 溢 (Overflow loss) increased significantly, while W 沉 (Sedimentation) increases due to high water levels in the pipeline network.

[0206] Based on comprehensive assessment, the main problems in the S1 zone are insufficient pipeline capacity leading to overflow during rainy days and siltation during dry days, and it should be given priority for inclusion in the key investigation list for combined rainwater and sewage connections.

[0207] (4) Diagnosis of S4 and S5 partitions (abnormalities in the normal area):

[0208] Although the load gap in areas S4 and S5 is negative ( M jThe ammonia nitrogen / BOD5 ratio was less than 0, seemingly indicating that the collection effect had "exceeded expectations." However, further analysis of water quality characteristics revealed that the ammonia nitrogen / BOD5 ratio was significantly low, deviating from the characteristic range of normal domestic sewage. Based on this, it was determined that the two areas did not have extremely high collection efficiency, but rather that there was intrusion of high-concentration industrial wastewater, leading to an abnormally high influent concentration.

[0209] (5) Diagnosis of S2, S3 and S6 zones: S2 load deficit is 147g / d, S3 load deficit is 24g / d and S6 load deficit is 160g / d. After analysis of water quality characteristic factors, no significant abnormality in conductivity or ammonia nitrogen / BOD5 ratio was found. It was determined that the main causes were local rainwater and sewage mixing or slight external water infiltration. It is recommended to include them in the routine investigation plan.

[0210] 7.4 Matching of governance measures based on diagnostic results;

[0211] Based on the above diagnostic conclusions, a precise governance plan will be matched from the measures database:

[0212] For zone S0 (first priority, dominated by groundwater infiltration): immediately conduct CCTV pipeline endoscopic inspection in this area, focusing on identifying pipeline damage points; implement trenchless repair (such as UV curing repair) to seal groundwater infiltration channels and restore the pipeline's airtightness.

[0213] For Zone S1 (second priority, dominated by hydraulic imbalance): CCTV inspections will be conducted to focus on identifying structural defects in the main pipeline; pipeline dredging and maintenance will be implemented to remove deposited pollutants. 沉 Optimize the scheduling of upstream pumping stations, lower the operating water level in the pipeline network, and reduce the risk of overflow.

[0214] For Zone S6 (third priority, localized rainwater and sewage mixing): it will be included in the next month's inspection plan, with priority given to investigating suspected rainwater and sewage mixing points; if necessary, localized rainwater and sewage separation renovations will be implemented.

[0215] For Zone S2 (fourth priority, mild rainwater and sewage mixing): it will be included in the next quarter's investigation plan and will be a key focus in daily inspections; if the monitoring load gap increases in the future, CCTV testing will be arranged.

[0216] For the S3 partition (fifth priority, basically meeting the standards): maintain regular monitoring, and do not arrange engineering measures for the time being.

[0217] For zones S4 and S5 (with negative load gaps and industrial wastewater intrusion): Although the total collection volume is "exceeding the limit", the water quality characteristic factors are abnormal, and it is necessary to conduct source tracing and investigation of wastewater dischargers; the focus should be on inspecting the pretreatment facilities of industrial enterprises and the situation of illegal discharge, and standardizing the discharge of industrial wastewater into the pipe network.

[0218] 8. Matching of engineering measures;

[0219] Based on the causes and priorities of anomalies in each analysis partition as determined in Section 7, this embodiment achieves precise conversion from assessment results to treatment plans by querying a pre-set database of wastewater system quality improvement and efficiency enhancement measures (as shown in Table 1).

[0220]

[0221] Table 1. Treatment Measures for Improving the Quality and Efficiency of Wastewater Systems

[0222] 8.1 Matching of targeted governance measures

[0223] Based on the specific diagnostic results of each zone, match the corresponding engineering and non-engineering measures, for example:

[0224] (1) For the S0 partition (first priority):

[0225] Diagnostic conclusion: Groundwater infiltration is dominant, electrical conductivity is low, and load gap is large.

[0226] Matching measures: Refer to the "Engineering Measures" category of "Rainwater and Sewage Separation, Pipeline Repair and Renovation" and the "Non-Engineering Measures" category of "Pipeline Inspection and Evaluation" in Table 1.

[0227] Specific plan: First, conduct pipeline network inspection and assessment (CCTV / QV inspection) to accurately locate the damage point; then, implement trenchless repair (such as ultraviolet light curing) to directly block the groundwater infiltration channel, reduce the amount of clean water infiltration, and increase the concentration of wastewater entering the plant.

[0228] (2) For the S1 partition (second priority):

[0229] Diagnostic conclusion: Hydraulic imbalance is the dominant factor, with overflow and sedimentation present.

[0230] Matching measures: Refer to the "Dredging and Maintenance" and "Sewage System Scheduling" categories in Table 1 under the "Non-Engineering Measures" category.

[0231] Specific plan: Implement dredging and maintenance to remove deposited pollutants from the pipeline network (reducing W). 沉 This increases the concentration of the transported liquid; at the same time, it optimizes the scheduling of upstream pumping station systems, lowers the operating liquid level in the pipeline network, and reduces the risk of overflow during rainy days.

[0232] 8.2 Coordination between the measure library and objective conditions

[0233] As shown in Table 1, in addition to precise measures for each zone, the impact of objective conditions must also be considered. If excessive rainfall occurs during the assessment period, leading to abnormal fluctuations in the load gap, the response strategies in Table 1 should be used to focus on changes in pollutant transport caused by rain overflow control and pipeline scouring. If necessary, the sampling frequency should be adjusted or correction coefficients should be calculated.

[0234] 8.3 After completing the above treatment measures, this embodiment will enter the next evaluation cycle (as shown in the following monthly cycle), and automatically reuse this method to recalculate the load gap and water quality characteristic factors of each zone:

[0235] Comparing key indicators before and after treatment, if the conductivity of zone S0 recovers to the characteristic range of domestic sewage (e.g., >800 μS / cm) and the load deficit per unit area ( MA j If the number of cases decreases significantly, the treatment is deemed effective.

[0236] Iterative optimization: If the indicators are not met, the closed-loop process of "monitoring-assessment-investigation-remediation-reassessment" will continue to be iterated until the wastewater system as a whole meets the assessment target for improving quality and efficiency.

[0237] As an optional optimization scheme, in order to further eliminate the accumulated bias in the calculation process and improve the evaluation accuracy, this embodiment also provides a modified model based on Long Short-Term Memory (LSTM) network.

[0238] Specifically, the pollutant load deficit calculated in the above steps is used as the input feature sequence of the LSTM model, and the measured efficiency improvement after the implementation of subsequent treatment measures is used as the output label for training. The trained LSTM model can learn and predict the nonlinear laws governing the change of pipeline network operating status over time, thereby dynamically correcting the load deficit value calculated based on topology analysis and outputting the corrected pollutant collection efficiency assessment results. The corrected assessment results can serve as a reliable basis for formulating pipeline network inspection and maintenance plans, helping operation and maintenance personnel to more confidently identify key areas.

[0239] This embodiment establishes a one-to-one correspondence between analysis zones and monitoring points, and analyzes them step by step from the end to the source along the upstream and downstream topological relationship model. It can separate the calculated wastewater flow and the actual total amount of pollutants discharged independently from each analysis zone from the mixed monitoring data, and realize the independent quantitative assessment of the pollutant collection efficiency at the zone level. This overcomes the shortcomings of existing methods that can only estimate the overall system and cannot accurately locate the problem zone.

[0240] This embodiment directly uses the measured flow data and pollutant concentration monitoring data of each monitoring point as the calculation input, and performs step-by-step deduction analysis along the pipeline topology, which reduces the calculation deviation caused by relying on theoretical sewage concentration, clean water concentration and other assumed parameters in the existing method, and makes the evaluation results closer to the actual operating state of the drainage system.

[0241] This embodiment calculates the pollutant load gap by comparing the actual total pollutant emissions of each analysis zone with the target total pollutant emissions, and determines the problem zones and their investigation priorities accordingly. This can intuitively reflect the actual contribution of each zone to the overall collection efficiency and provide a basis for the precise deployment of geophysical detection and remediation projects.

[0242] This embodiment introduces rainfall data in the data preprocessing stage, divides the assessment period into dry and rainy periods, calculates the actual total pollutant emissions of each analysis zone under the two operating conditions, and quantifies the degree of disturbance of rainfall to sewage collection efficiency by comparing the ratio of total pollutant emissions during rainy and dry days. This can directly reflect the impact of combined sewer overflow and overflow pollution on the sewage collection rate of the area, and provide a direct basis for the investigation of combined sewer overflow and the control of combined sewer overflow.

[0243] In this embodiment, when constructing the upstream and downstream topological relationship model, a weighted pipe network hydraulic model is further constructed. Physical attribute data such as pipe diameter, pipe length, and pipe material roughness are embedded into the model as weighting factors. A pre-trained convolutional neural network model is used to identify the pipe endoscopic image to obtain the pipe siltation coverage rate and structural defect coefficient. Based on the above coefficients, the effective flow cross-sectional area and pipe material roughness are dynamically corrected, which can eliminate the deviation caused by factors such as pipe siltation and structural defects in the flow analysis calculation and improve the accuracy of flow and pollutant analysis results.

[0244] Based on the calculation of pollutant load gap, this embodiment introduces water quality characteristic factors such as ammonia nitrogen concentration, total phosphorus concentration, and conductivity, and establishes anomaly judgment rules in combination with load gap values. Based on clear quantitative indicators, it can quickly identify different abnormal operating conditions. For areas with more serious problems or complex causes, this embodiment further introduces flow conservation equations and pollutant mass conservation equations under abnormal operating conditions. Through qualitative or semi-quantitative analysis, it infers the main causes of abnormal operating conditions and selects corresponding treatment measures from a preset measure library accordingly.

[0245] This embodiment constructs a multi-dimensional load gap density index system, including load gap per unit area, load gap per unit pipeline length, and load gap per unit operating water depth at full capacity. Based on the index values, the problem investigation priority of each analysis zone is ranked, which can comprehensively reflect the pollution load density, pipeline distribution characteristics, and operating conditions of different zones, and provide priority guidance for governance measures such as CCTV pipeline endoscopic inspection, pipeline defect repair, and pump station scheduling optimization.

[0246] The entire method described in this embodiment can be deployed on a smart water management platform, enabling fully automated processing from data retrieval, topology modeling, flow and pollutant analysis, load gap calculation to priority ranking and treatment suggestion generation. It also supports routine assessments on a monthly or quarterly basis and allows for historical result retrospective analysis. After the pipeline network renovation is completed, this method can be reused for closed-loop assessments to quantify changes in sewage collection efficiency before and after treatment, providing technical support for the long-term operation and dynamic optimization of urban sewage systems.

[0247] Example 3:

[0248] Finally, this application also proposes a computing device, which includes a processor and a memory. The memory stores computer programs, and the processor executes the instructions stored in the memory so that the computer device performs a method for evaluating the quality and efficiency improvement effect of a wastewater system based on water quality and quantity, as described in the above embodiments.

[0249] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for evaluating the effectiveness of wastewater system quality and efficiency improvement based on water quality and quantity, characterized in that, include: S1. Obtain multi-source data on the sewage collection service area covered by a sewage treatment plant and its supporting pipe network, including the topology of the drainage pipe network, flow direction data and distribution of monitoring points; Based on the drainage network topology, monitoring point distribution, and the network confluence boundary and pump station service area, each analysis zone is divided level by level, and the correspondence between each analysis zone and each monitoring point is established; wherein, the network confluence boundary and pump station service area are determined analytically based on the drainage network topology. S2. Based on the drainage network topology and flow direction data, construct an upstream and downstream topological relationship model between the monitoring points; obtain the measured flow data of each monitoring point, and along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point, analyze the measured flow of each monitoring point into the calculated sewage flow of each analysis zone step by step. S3. Obtain pollutant concentration monitoring data at each monitoring point, combine the calculated wastewater flow rate of each analysis zone with the upstream and downstream topological relationship model, and analyze and calculate the actual total pollutant discharge of each analysis zone step by step along the upstream and downstream topological relationship model from the end monitoring point to the source monitoring point. S4. Multiply a preset target concentration of pollutants by the calculated wastewater flow rate of each analysis zone to obtain the total target discharge of pollutants for each analysis zone. S5. Compare the actual total pollutant emissions of each analysis zone with the target total pollutant emissions, calculate the pollutant load gap value, and determine the problem zones with insufficient pollutant collection and their investigation priority ranking in each analysis zone based on the pollutant load gap value.

2. The method according to claim 1, characterized in that, The wastewater collection service area is based on the wastewater treatment plant as the main water catchment node and assessment benchmark. The analysis zones are drainage sub-regions divided upstream of the wastewater treatment plant based on the pipe network confluence boundary and the service range of the pumping station; the boundaries between each analysis zone are consistent with the pipe network confluence path, and the wastewater in each analysis zone flows out through the corresponding monitoring point; The monitoring point for each analysis zone is located in the inspection well before the main pipe of that analysis zone connects to the next level of the pipeline network; the water quantity and water quality data measured at each monitoring point are the total water quantity and total water quality after mixing the water from that monitoring point and all upstream analysis zones.

3. The method according to claim 1, characterized in that, When obtaining the measured flow data of each monitoring point in S2, the method further includes: obtaining the rainfall data of the sewage collection service area within a preset assessment period, and dividing the time period within the preset assessment period into dry period and rainy period according to the rainfall data; In S3, when analyzing and calculating the actual total pollutant emissions of each analysis zone step by step, the measured flow data and pollutant concentration monitoring data corresponding to the dry weather period and the rainy weather period are used respectively to calculate the actual total pollutant emissions of each analysis zone under dry weather conditions and rainy weather conditions. In step S5, when determining the problem zones with insufficient pollutant collection and their investigation priorities in each analysis zone based on the pollutant load gap value, the actual total amount of pollutant emissions under dry and rainy conditions is compared to quantify the impact of rainfall on the pollutant collection efficiency of each analysis zone.

4. The method according to claim 1, characterized in that, When constructing the upstream and downstream topological relationship model between the monitoring points in S2, it also includes: Obtain the physical property data of each pipe section, including pipe diameter, pipe length, pipe material roughness, and effective flow cross-sectional area calculated based on pipe diameter; The physical attribute data is embedded as a weighting factor into the topological relationship model to construct a weighted network hydraulic model; Acquire historical or real-time pipeline endoscopic image data corresponding to each analysis partition; The pre-trained convolutional neural network model is used to identify the endoscopic images of the pipe to determine the siltation coverage and structural defect coefficient inside the pipe. Based on the siltation coverage rate and structural defect coefficient, the effective flow cross-sectional area and pipe roughness in the physical attribute data are dynamically corrected, and the corrected physical attribute data is embedded in the weighted pipe network hydraulic model for weight allocation correction of the measured flow data at each monitoring point.

5. The method according to claim 1, characterized in that, The step-by-step calculation of wastewater flow rate for each analysis zone in S2 follows the following flow balance relationship: ； in, For analyzing partition S j Calculate the wastewater flow rate, Q j Let D(j) be the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, and Q be the measured flow rate of node j. k This represents the measured flow rate at upstream monitoring node k.

6. The method according to claim 1, characterized in that, The step-by-step analysis and calculation of the actual total pollutant emissions for each analysis zone in S3 follows the following pollutant mass balance relationship for any monitoring node j in the pipeline network: ； Among them, M j For analyzing partition S j The actual total amount of pollutants emitted, C j For monitoring pollutant concentration data at monitoring node j, Q j Let D(j) be the set of monitoring nodes directly upstream of node j in the upstream-downstream topology model, and C be the measured flow rate of node j. k Q represents the pollutant concentration monitoring data at upstream monitoring node k. k This represents the measured flow rate at upstream monitoring node k.

7. The method according to claim 1, characterized in that, In step S5, based on the pollutant load deficit value, the problem zones with insufficient pollutant collection in each of the analysis zones and their investigation priority ranking are determined, including: Calculate the following density indices for each analysis partition, and prioritize the issues in each analysis partition based on these density indices: Load gap per unit area MA j The calculation formula is: MA j = M j / A j ; Load gap per unit pipeline length ML j The calculation formula is: ML j = M j / L j ; Full-pipe unit operating water depth load gap M Hj The calculation formula is: M Hj = M j / H j ; in, M j For analyzing partition S j The pollutant load deficit value, A j For analyzing partition S j area, L j For analyzing partition S j The length of the sewage pipe network within the area, H j For analyzing partition S j The average operating water depth in a fully operational sewage pipe network; The higher the value of the density index, the higher the priority of the corresponding analysis and screening.

8. The method according to claim 1, characterized in that, Following S5, the following is also included: Obtain water quality characteristic factor data for each monitoring point, wherein the water quality characteristic factor includes at least one of ammonia nitrogen concentration, total phosphorus concentration, and conductivity. Based on the water quality characteristic factor data and the actual total amount of pollutants discharged in each analysis zone, the composition of pollution sources in each analysis zone is analyzed to identify at least one abnormal situation, such as leakage of domestic sewage, intrusion of industrial wastewater, or infiltration of groundwater. When the actual total pollutant discharge of the analysis zone is lower than the target total pollutant discharge of the analysis zone, and the conductivity is lower than the preset groundwater characteristic threshold, it is determined that there is groundwater infiltration in the analysis zone. When the actual total pollutant discharge of the analysis zone is higher than the target total pollutant discharge of the analysis zone, and the combination of water quality characteristic factors is abnormal, it is determined that there is industrial wastewater intrusion in the analysis zone.

9. The method according to claim 1, characterized in that, Following S5, it also includes: Based on the pre-constructed flow conservation equation and pollutant mass conservation equation under abnormal operating conditions, combined with the pollutant load deficit value and the water quality characteristic factor data, the main causes of abnormal operating conditions in each analysis zone are analyzed, and corresponding treatment measures are selected from the preset measure library accordingly.

10. The method according to claim 9, characterized in that, The flow conservation equation is: ； The mass conservation equation for the pollutants is: ； Where Q1 and C1 are the measured wastewater volume and measured pollutant concentration at the end of the drainage section under abnormal operating conditions, respectively; Q0 ​​and C0 are the measured wastewater flow rate and measured pollutant concentration at the upstream input section, respectively; Q 清 C 清 These represent the volume and concentration of clean water flowing into the sewage system under abnormal operating conditions caused by mixed rainwater and sewage connections; Q 地 C 地 These represent the amount of groundwater infiltrating at the location of pipeline rupture under abnormal operating conditions and the background concentration of groundwater; Q 调 C 调 These represent the volume of overflow wastewater collected and returned to the wastewater system by the storage facilities and its average pollutant concentration; Q 渗i C 渗i These represent the volume of sewage leaking from the i-th pipe rupture under abnormal operating conditions and the sewage concentration at the leak point; Q 溢j C 溢j These represent the overflow sewage volume and sewage concentration at the j-th location under abnormal operating conditions caused by combined sewer overflow or pipeline overload; W 沉 This refers to the amount of insoluble pollutants deposited in the pipes during high-water-level operation of the pipeline network under abnormal operating conditions; W 消 This represents the amount of soluble pollutants lost through biodegradation during pipeline transportation; n represents the total number of sewage seepage points in the pipeline network, and m represents the total number of overflow points in the pipeline network.

Images