Dynamic hydraulic simulation and emergency cooperative fault early warning disposal method for drainage pipe network

Abstract

The invention discloses a dynamic hydraulic simulation and emergency cooperation fault early warning processing method for a drainage pipe network, relates to the technical field of intelligent drainage pipe network operation and maintenance, and solves the technical problems that monitoring lags behind, diagnosis depends on manpower, and department response cooperation is insufficient in a traditional method. By constructing a dynamic risk early warning model, real-time accurate early warning of faults such as pipe network blockage and leakage is realized; the AI model is utilized to automatically diagnose fault root causes and intelligently match an optimal disposal scheme, and a mode depending on artificial experience is changed; emergency resource cooperative scheduling is carried out by adopting an improved algorithm, and multiple departments such as traffic and environmental protection are linked, so that the cross-department cooperation efficiency is effectively improved; and finally, feeding back a closed loop through a disposal result, and continuously optimizing an early warning model and a scheduling strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent drainage pipe network operation and maintenance, in particular to a drainage pipe network dynamic hydraulic simulation and emergency collaborative fault early warning and disposal method. BACKGROUND

[0002] As a core component of urban infrastructure, the operation state of urban drainage pipe network directly affects the urban flood control safety and water environment quality. In the method of "a drainage pipe network defect hidden danger risk assessment method, electronic equipment and storage medium", based on the static data set of the pipe network, a drainage pipe network defect hidden danger risk variable factor evaluation model is constructed, but there are still the following key problems:

[0003] Early warning is static: the existing evaluation is mostly based on historical data and periodic detection results, and lacks dynamic fusion of real-time hydraulic parameters, which cannot capture the risk changes in dynamic scenarios such as rainfall and sudden increase of pipe network load in real time, resulting in delayed early warning and missed optimal disposal opportunity;

[0004] Disposal fragmentation: after the fault occurs, emergency resources such as personnel and equipment need to be matched, and disposal schemes need to be developed, and the cross-departmental collaborative efficiency is low, and the disposal process involves operation and maintenance units, transportation departments, flood control offices and other departments. Especially in the multi-fault point concurrent scene, it is easy to cause resource allocation confusion and prolong the fault disposal time;

[0005] No feedback optimization mechanism: the fault disposal result is not linked with the early warning model, and the early warning threshold and resource scheduling strategy cannot be updated according to the actual disposal effect, resulting in gradual decline of model accuracy with changes in pipe network operation conditions, and difficulty in adapting to long-term operation and maintenance requirements.

[0006] In view of the above problems, an integrated system integrating dynamic early warning, intelligent diagnosis, collaborative scheduling and feedback optimization is proposed to solve the defects of "heavy evaluation, light disposal", "heavy static, light dynamic" in the prior art, and to improve the drainage pipe network fault response efficiency and operation and maintenance intelligent level. SUMMARY

[0007] In order to solve the technical problems of monitoring lag, diagnosis relying on manual work and insufficient department response collaboration in the traditional method, the present application provides a drainage pipe network dynamic hydraulic simulation and emergency collaborative fault early warning and disposal method, comprising:

[0008] S1: multi-source data acquisition and preprocessing, including acquiring the basic static data, real-time dynamic data and historical data of the drainage pipe network, and preprocessing the data to construct a space-time-attribute three-dimensional database;

[0009] S2: dynamic risk early warning model construction, calculating the dynamic risk value based on the space-time-attribute three-dimensional database, and dividing the early warning level according to the dynamic risk value;

[0010] S3: AI fault diagnosis and handling solution generation. It uses LSTM and hydraulic simulation coupled model to diagnose fault types and root causes, and automatically matches and generates emergency handling solutions.

[0011] S4: Emergency resource collaborative scheduling, using an improved Dijkstra algorithm to calculate the optimal scheduling path for emergency resources, and realizing cross-departmental linkage through a collaborative scheduling platform;

[0012] S5: Feedback on handling results and model optimization, updating the early warning model and scheduling strategy based on the fault handling results.

[0013] Furthermore, S1 includes:

[0014] S1.1: Collect multi-source data of the drainage pipe network, including: basic static data: pipe network GIS data, including: pipe diameter, material, burial depth, and topological connection relationship; emergency resource data: location of pipe network dredging and sealing equipment, number and skill level of maintenance personnel, and emergency material reserve points; real-time dynamic data: real-time flow (Q), pressure (P), and water level (H) data collected by ultrasonic flow meters, pressure sensors, and level gauges deployed at key nodes of the pipe network, with a sampling frequency of ≥1 time / minute; real-time rainfall intensity (R) data obtained from meteorological stations; historical data: pipe network fault records for the past 5 years, including type, location, treatment plan, degree of loss, and historical fluctuation data of hydraulic parameters;

[0015] S1.2: Preprocessing of multi-source data: using 3 The criteria remove outliers from real-time data and use a Kalman filter algorithm to fuse multi-sensor data to reduce environmental interference.

[0016] S1.3: Associate the preprocessed static and dynamic data with the pipeline GIS map, import them into Esri ArcGIS software and match them with GeoTIFF format DOM data, set spatial reference and projection parameters to ensure topological overlay, and construct the spatial-temporal-attribute three-dimensional database.

[0017] Furthermore, S2 includes:

[0018] S2.1: Construct the dynamic risk value calculation formula, the expression is:

[0019]

[0020] in, This is a dynamic risk value;

[0021] The probability of a fault occurring in real time is set to a value of 1-5. It is calculated by coupling the real-time hydraulic anomaly level, such as the proportion of flow fluctuations exceeding the threshold and the magnitude of pressure drops, with historical fault frequencies. The formula is as follows:

[0022]

[0023] in, The rated flow rate is ΔQ, where ΔQ is the difference between the real-time flow rate and the rated flow rate.

[0024] To indicate the real-time severity of the fault, a value of 1-5 is used, taking into account the impact on population density, estimated economic losses, and the scope of environmental risks. The multivariate weighted algorithm is used (weight allocation: population density 30%, economic losses 30%, environmental scope 40%).

[0025] K is the pipeline load factor, ranging from 1.0 to 2.0. When the rainfall intensity R ≥ 50 mm / 24h, K = 1.8.

[0026] When 50mm / 24h > R ≥ 20mm / 24h, K = 1.4; when R < 20mm / 24h, K = 1.0.

[0027] S2.2 Based on dynamic risk values, early warning levels are divided to form a correspondence between early warning levels and response requirements.

[0028] Furthermore, S3 includes:

[0029] S3.1 Construct a coupled LSTM and hydraulic simulation model, input real-time hydraulic data and historical fault samples, and automatically diagnose fault types and root causes;

[0030] S3.2 Establish a fault handling plan matching database, automatically match and generate emergency handling plans based on fault type, location, and scope of impact, and specify the required equipment model, number of personnel, and operation time.

[0031] Furthermore, S4 includes:

[0032] S4.1 employs an improved Dijkstra algorithm, combined with real-time traffic data, to calculate the optimal emergency resource scheduling path. The objective function is:

[0033]

[0034] In the formula, Let be the objective function. For equipment operating time, For the time of personnel assembly, Preset processing time;

[0035] S4.2 builds a cross-departmental collaborative scheduling platform, which connects operation and maintenance units, transportation departments, and environmental protection departments through API interfaces to push traffic control instructions, water quality monitoring tasks, and resource status information.

[0036] Furthermore, S5 includes:

[0037] After the S5.1 fault handling is completed, the handling results are entered, including the blockage removal rate, leakage repair effect, and handling time. The handling compliance rate is calculated, and the compliance standards are: blockage removal rate ≥95% and pressure recovery after leakage repair ≥90%.

[0038] S5.2 If the handling compliance rate is ≥90%, the fault data will be added to the model training set as positive samples; if the compliance rate is <90%, the weight parameters of the dynamic risk value calculation formula and the resource matching logic will be adjusted to ensure that the model accuracy improves by ≥3% every quarter.

[0039] An electronic device is also provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the drainage network fault early warning and handling method based on dynamic hydraulic simulation and emergency resource coordination as described above.

[0040] A computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, provides the aforementioned method for early warning and handling of drainage network faults based on dynamic hydraulic simulation and emergency resource coordination.

[0041] The beneficial effects of this invention are:

[0042] This invention constructs a dynamic risk early warning model by integrating real-time hydraulic data and historical fault data to achieve real-time and accurate early warning of faults; it uses an AI model to automatically diagnose fault types and generate optimal handling solutions, intelligently dispatches emergency resources from multiple departments, and improves collaborative efficiency; at the same time, it establishes a handling result feedback mechanism to continuously optimize the early warning model and dispatching strategy, forming an operation and maintenance closed loop, and improving the intelligent level of drainage pipe network operation and maintenance and emergency response capabilities. Attached Figure Description

[0043] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0044] The technical solution of the present invention will be further described below with reference to embodiments, but it is not limited thereto. Any modifications or equivalent substitutions to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be covered within the protection scope of the present invention. In the following embodiments, the process equipment or apparatus not specifically specified are all conventional equipment or apparatus in the art. Unless otherwise specified, the raw materials used in the embodiments of the present invention are all commercially available; unless otherwise specified, the technical means used in the embodiments of the present invention are all conventional means well known to those skilled in the art.

[0045] Example 1, combined with Figure 1 This embodiment describes a method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks, including: S1: multi-source data acquisition and preprocessing, including acquiring basic static data, real-time dynamic data and historical data of the drainage pipe network, and preprocessing the data to construct a three-dimensional database of space-time-attribute;

[0046] S2: Construction of dynamic risk early warning model, calculating dynamic risk value based on the space-time-attribute three-dimensional database, and classifying early warning level according to the dynamic risk value;

[0047] S3: AI fault diagnosis and handling solution generation. It uses LSTM and hydraulic simulation coupled model to diagnose fault types and root causes, and automatically matches and generates emergency handling solutions.

[0048] S4: Emergency resource collaborative scheduling, using an improved Dijkstra algorithm to calculate the optimal scheduling path for emergency resources, and realizing cross-departmental linkage through a collaborative scheduling platform;

[0049] S5: Feedback on handling results and model optimization, updating the early warning model and scheduling strategy based on the fault handling results.

[0050] Furthermore, S1 includes:

[0051] S1.1: Collect multi-source data of the drainage pipe network, including: basic static data: pipe network GIS data, including: pipe diameter, material, burial depth, and topological connection relationship; emergency resource data: location of pipe network dredging and sealing equipment, number and skill level of maintenance personnel, and emergency material reserve points; real-time dynamic data: real-time flow (Q), pressure (P), and water level (H) data collected by ultrasonic flow meters, pressure sensors, and level gauges deployed at key nodes of the pipe network, with a sampling frequency of ≥1 time / minute; real-time rainfall intensity (R) data obtained from meteorological stations; historical data: pipe network fault records for the past 5 years, including type, location, treatment plan, degree of loss, and historical fluctuation data of hydraulic parameters;

[0052] S1.2: Preprocessing of multi-source data: using 3 The criteria remove outliers from real-time data and use a Kalman filter algorithm to fuse multi-sensor data to reduce environmental interference.

[0053] S1.3: Associate the preprocessed static and dynamic data with the pipeline GIS map, import them into Esri ArcGIS software and match them with GeoTIFF format DOM data, set spatial reference and projection parameters to ensure topological overlay, and construct the spatial-temporal-attribute three-dimensional database.

[0054] Furthermore, S2 includes:

[0055] S2.1: Construct the dynamic risk value calculation formula, the expression is:

[0056]

[0057] in, This is a dynamic risk value;

[0058] The probability of a fault occurring in real time is set to a value of 1-5. It is calculated by coupling the real-time hydraulic anomaly level, such as the proportion of flow fluctuations exceeding the threshold and the magnitude of pressure drops, with historical fault frequencies. The formula is as follows:

[0059]

[0060] in, The rated flow rate is ΔQ, where ΔQ is the difference between the real-time flow rate and the rated flow rate.

[0061] To indicate the real-time severity of the fault, a value of 1-5 is used, taking into account the impact on population density, estimated economic losses, and the scope of environmental risks. The multivariate weighted algorithm is used (weight allocation: population density 30%, economic losses 30%, environmental scope 40%).

[0062] K is the pipeline load factor, ranging from 1.0 to 2.0. When the rainfall intensity R ≥ 50 mm / 24h, K = 1.8.

[0063] When 50mm / 24h > R ≥ 20mm / 24h, K = 1.4; when R < 20mm / 24h, K = 1.0.

[0064] S2.2 Based on dynamic risk values, early warning levels are divided to form a correspondence between early warning levels and response requirements.

[0065] Specifically, the warning level classification table is shown below:

[0066]

[0067] Furthermore, S3 includes:

[0068] S3.1 Construct a coupled LSTM and hydraulic simulation model, input real-time hydraulic data and historical fault samples, and automatically diagnose fault types and root causes;

[0069] Specifically, in S3.1:

[0070] If the flow rate drops by ≥30% and the pressure rises by ≥20%, it is determined to be "pipeline blockage". Further, the location of the blockage is located by using CCTV image data from inside the pipe.

[0071] If the pressure drops by ≥30% and the water level drops by ≥15%, it is determined to be a "pipeline leak". The leak range is then determined by combining the soil moisture sensor data.

[0072] If there is no response in the flow or pressure of a local area, it is determined to be a "valve failure", and the failure type is confirmed by feedback of the valve control signal.

[0073] S3.2 Establish a fault handling plan matching database, automatically match and generate emergency handling plans based on fault type, location, and scope of impact, and specify the required equipment model, number of personnel, and operation time.

[0074] Specifically, the S3.2 treatment plan is automatically generated as shown in the table below:

[0075]

[0076] Furthermore, S4 includes:

[0077] S4.1 employs an improved Dijkstra algorithm, combined with real-time traffic data, to calculate the optimal emergency resource scheduling path. The objective function is:

[0078]

[0079] In the formula, Let be the objective function. For equipment operating time, For the time of personnel assembly, Preset processing time;

[0080] Specifically, a traffic congestion coefficient λ is introduced, where λ = 1.0-2.5. The higher the congestion index, the larger λ becomes, and the travel time is corrected as follows:

[0081]

[0082] in, This refers to the travel time under conditions of no traffic congestion.

[0083] S4.2 builds a cross-departmental collaborative scheduling platform, which connects operation and maintenance units, transportation departments, and environmental protection departments through API interfaces to push traffic control instructions, water quality monitoring tasks, and resource status information.

[0084] Specifically, S4.2 is as follows:

[0085] Build a collaborative scheduling platform to connect operation and maintenance units, transportation departments, and environmental protection departments through API interfaces:

[0086] Send a temporary traffic control order to the transportation department within a 500m radius of the fault location;

[0087] The water quality monitoring data of the leak was sent to the environmental protection department, and the protection plan for the surrounding water bodies was activated simultaneously.

[0088] It displays the location and progress of resources in real time and supports online communication among multiple departments.

[0089] Furthermore, S5 includes:

[0090] After the S5.1 fault handling is completed, the handling results are entered, including the blockage removal rate, leakage repair effect, and handling time. The handling compliance rate is calculated, and the compliance standards are: blockage removal rate ≥95% and pressure recovery after leakage repair ≥90%.

[0091] S5.2 If the handling compliance rate is ≥90%, the fault data will be added to the model training set as positive samples; if the compliance rate is <90%, the weight parameters of the dynamic risk value calculation formula and the resource matching logic will be adjusted to ensure that the model accuracy improves by ≥3% every quarter.

[0092] Specifically, the model parameter adjustment rule is as follows: if the compliance rate for handling the same type of fault is less than 90% for three consecutive times, then... The weighting of the degree of hydraulic anomaly has been increased from 60% to 65%-70%, while the equipment dispatch radius has been reduced to ≤1.5km.

[0093] Example 2 illustrates the complete operation process of a system by taking the handling of a DN800 drainage pipeline failure in the old city area of ​​a certain city as an example.

[0094] S1. Data Acquisition and Preprocessing:

[0095] Ultrasonic flow meters and pressure sensors were deployed at key nodes of the drainage network to collect real-time data on flow rate Q=120m³ / h (rated flow rate 200m³ / h, ΔQ=-40%) and pressure P=0.3MPa. Simultaneously, rainfall intensity R=65mm / 24h was acquired (K=1.8). The data, after Kalman filtering, was entered into a three-dimensional database.

[0096] S2. Dynamic Risk Warning:

[0097] calculate =0.6×(40%)+0.4×(Fault frequency of this node in the past year: 3 times / monitoring duration: 8760h)=0.6×0.4+0.4×0.00034=0.2401, take the value 2); Based on population density (800 people within 500m, score 4), economic loss (15 shops in the vicinity, score 4), and environmental scope (adjacent to a river, score 5), the calculation is: 0.3×4+0.3×4+0.4×5=4.4, and the value is 4. =2×4×1.8=14.4, which is determined to be a Level 2 warning. The system automatically pushes the warning information to the operation and maintenance manager.

[0098] S3. Fault Diagnosis and Solution Generation:

[0099] Using LSTM and hydraulic simulation model inputs Q and P data, and comparing with historical samples, it was determined to be "pipeline leakage". Combined with soil moisture sensor data, the leakage point was located at the intersection of A Road and B Street. Automatic treatment plan was generated: matching pressurized sealing tools, 2 maintenance personnel, preset treatment time of 2 hours, and simultaneously starting water quality monitoring of surrounding rivers.

[0100] S4. Resource Coordination Scheduling:

[0101] Using an improved Dijkstra algorithm combined with real-time traffic data: Road A has a congestion index of 1.2, the equipment scheduling route is calculated as follows: from the reserve point to B Street via Expressway C, with a travel time of 22 minutes; the personnel assembly point is set as a parking lot within 100m of the fault point, with an assembly time of 15 minutes; the platform automatically pushes temporary control instructions to the traffic department, determining the control area as 300m around the fault point, and pushes water quality monitoring tasks to the environmental protection department.

[0102] S5. Feedback and Optimization:

[0103] After the treatment was completed, the system recorded a blockage clearance rate of 98%, pressure recovery to 0.42 MPa, and a treatment compliance rate of 95%. The system used this data as a positive sample to supplement the model and update it. The weight of the degree of hydraulic anomaly was increased from 60% to 65%, and the optimized model's diagnostic response time for similar leakage faults was shortened to 8 seconds.

[0104] This invention automates the entire process of "early warning-diagnosis-schedule-handling-feedback" for drainage network faults through dynamic hydraulic simulation and emergency resource coordination. It reduces the fault response time to one-third of the traditional manual method and increases the handling compliance rate to over 92%, effectively reducing losses such as waterlogging and water pollution caused by network faults, and providing technical support for the operation and maintenance of smart drainage networks.

Claims

1. A method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks, characterized in that, include: S1: Multi-source data acquisition and preprocessing, including acquiring basic static data, real-time dynamic data and historical data of drainage pipe network, and preprocessing the data to construct a spatial-temporal-attribute three-dimensional database; S2: Construction of dynamic risk early warning model, calculating dynamic risk value based on the space-time-attribute three-dimensional database, and classifying early warning level according to the dynamic risk value; S3: AI fault diagnosis and handling solution generation. It uses LSTM and hydraulic simulation coupled model to diagnose fault types and root causes, and automatically matches and generates emergency handling solutions. S4: Emergency resource collaborative scheduling, using an improved Dijkstra algorithm to calculate the optimal scheduling path for emergency resources, and realizing cross-departmental linkage through a collaborative scheduling platform; S5: Feedback on handling results and model optimization, updating the early warning model and scheduling strategy based on the fault handling results.

2. The method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks according to claim 1, characterized in that, S1 includes: S1.1: Collect multi-source data of the drainage pipe network, including: basic static data: pipe network GIS data, including: pipe diameter, material, burial depth, and topological connection relationship; emergency resource data: location of pipe network dredging and sealing equipment, number and skill level of maintenance personnel, and emergency material reserve points; real-time dynamic data: real-time flow, pressure, and water level data collected by ultrasonic flow meters, pressure sensors, and level gauges deployed at key nodes of the pipe network, with a sampling frequency of ≥1 time / minute; real-time rainfall intensity data obtained from meteorological stations; historical data: pipe network fault records for the past 5 years, including type, location, treatment plan, degree of loss, and historical fluctuation data of hydraulic parameters. S1.2: Preprocessing of multi-source data: using 3 The criteria remove outliers from real-time data and use a Kalman filter algorithm to fuse multi-sensor data to reduce environmental interference. S1.3: Associate the preprocessed static and dynamic data with the pipeline GIS map, import them into Esri ArcGIS software and match them with GeoTIFF format DOM data, set spatial reference and projection parameters to ensure topological overlay, and construct the spatial-temporal-attribute three-dimensional database.

3. The method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks according to claim 1, characterized in that, S2 includes: S2.1: Construct the dynamic risk value calculation formula, the expression is: in, This is a dynamic risk value; The probability of a fault occurring in real time is set to a value of 1-5. It is calculated by coupling the real-time hydraulic anomaly level, such as the proportion of flow fluctuations exceeding the threshold and the magnitude of pressure drops, with historical fault frequencies. The formula is as follows: in, The rated flow rate is ΔQ, where ΔQ is the difference between the real-time flow rate and the rated flow rate. To determine the real-time severity of the fault, a value of 1-5 is used, taking into account the impact on population density, estimated economic losses, and the scope of environmental risks, and a multivariate weighted algorithm is adopted. K is the pipeline load factor, ranging from 1.0 to 2.

0. When the rainfall intensity R ≥ 50 mm / 24h, K = 1.

8. When 50mm / 24h > R ≥ 20mm / 24h, K = 1.4; when R < 20mm / 24h, K = 1.

0. S2.2 Based on dynamic risk values, early warning levels are divided to form a correspondence between early warning levels and response requirements.

4. The method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks according to claim 1, characterized in that, S3 includes: S3.1 Construct a coupled LSTM and hydraulic simulation model, input real-time hydraulic data and historical fault samples, and automatically diagnose fault types and root causes; S3.2 Establish a fault handling plan matching database, automatically match and generate emergency handling plans based on fault type, location, and scope of impact, and specify the required equipment model, number of personnel, and operation time.

5. The method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks according to claim 1, characterized in that, S4 includes: S4.1 employs an improved Dijkstra algorithm, combined with real-time traffic data, to calculate the optimal emergency resource scheduling path. The objective function is: In the formula, Let be the objective function. For equipment operating time, For the time of personnel assembly, Preset processing time; S4.2 builds a cross-departmental collaborative scheduling platform, which connects operation and maintenance units, transportation departments, and environmental protection departments through API interfaces to push traffic control instructions, water quality monitoring tasks, and resource status information.

6. The method for dynamic hydraulic simulation and emergency collaborative fault early warning and handling of drainage pipe networks according to claim 1, characterized in that, S5 includes: After the S5.1 fault handling is completed, the handling results are entered, including the blockage removal rate, leakage repair effect, and handling time. The handling compliance rate is calculated, and the compliance standards are: blockage removal rate ≥95% and pressure recovery after leakage repair ≥90%. S5.2 If the handling compliance rate is ≥90%, the fault data will be added to the model training set as positive samples; if the compliance rate is <90%, the weight parameters of the dynamic risk value calculation formula and the resource matching logic will be adjusted to ensure that the model accuracy improves by ≥3% every quarter.

7. An electronic device, characterized in that, The device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the drainage network fault early warning and handling method based on dynamic hydraulic simulation and emergency resource coordination as described in any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the drainage network fault early warning and handling method based on dynamic hydraulic simulation and emergency resource coordination as described in any one of claims 1-6.

Images