CN-121980801-A - Method, device, equipment and medium for controlling leakage damage of water supply pipe network

CN121980801ACN 121980801 ACN121980801 ACN 121980801ACN-121980801-A

Abstract

The invention provides a water supply pipe network leakage damage management and control method, device, equipment and medium, which comprise the steps of obtaining characteristic data representing the running state of a pipe network, constructing a multi-scale coupling digital twin model capable of dynamically mapping and simulating the running state of the pipe network through space-time correlation fusion, obtaining corresponding fusion data, identifying the leakage damage state in the water supply pipe network based on the multi-scale coupling digital twin model and the fusion data thereof, outputting the leakage damage evolution trend and the risk level of the leakage damage state through a time sequence prediction model, solving the optimal solution through a multi-objective optimization algorithm, carrying out simulation verification through the multi-scale coupling digital twin model, generating a final loss control strategy, and issuing a control instruction and a maintenance task based on the final loss control strategy. Compared with the prior art, the method and the system reduce the damage degree of the water supply network leakage by identifying the water supply network leakage state, actively predicting the evolution trend and generating the optimal solution loss control strategy.

Inventors

HUANG JIANRU
LIN KE
KONG YONGPING
JIA YUXIN

Assignees

天翼物联科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260128

Claims (10)

1. The utility model provides a water supply pipe network leaks damage control method which characterized in that includes: acquiring acoustic signals, water pressure data and flow data of a water supply network, and performing feature extraction on the acoustic signals, the water pressure data and the flow data to obtain feature data representing the running state of the network; based on the spatial structure data, pipeline materials, pipe age history data and soil corrosion environment parameters of the water supply network and the characteristic data, constructing a multi-scale coupling digital twin model capable of dynamically mapping and simulating the running state of the network through space-time correlation fusion, and obtaining corresponding fusion data; identifying a leakage state in the water supply network based on the multi-scale coupling digital twin model and fusion data thereof, and outputting a leakage evolution trend and a risk level of the leakage state through a time sequence prediction model; Based on the leakage evolution trend and the risk level, taking water saving benefits, the influence of water pressure at a user side and operation cost as optimization targets, solving an optimal solution by utilizing a multi-target optimization algorithm, and performing simulation verification through the multi-scale coupling digital twin model to generate a final loss control strategy; And based on the final loss control strategy, issuing a control instruction to a valve execution unit of the water supply network, and simultaneously pushing maintenance task information to an operation and maintenance system.
2. The method for controlling leakage and damage of a water supply pipe network according to claim 1, wherein the constructing a multi-scale coupled digital twin model capable of dynamically mapping and simulating the running state of the pipe network by space-time correlation fusion based on the spatial structure data, pipe material, pipe age history data and soil corrosion environment parameters of the water supply pipe network and the characteristic data, and obtaining corresponding fusion data comprises the following steps: calling a design drawing of the water supply pipe network and geographic information system data, and constructing a geometric mapping model containing geometric shapes and spatial position information of pipelines, valves and pumping stations in a three-dimensional graphic engine; Acquiring pipeline material parameters, pipe wall thickness and pipe age historical data of the water supply network, and calculating to obtain an aging rate and a residual life evaluation value of the pipeline through a material corrosion dynamics model by combining the soil corrosion environment parameters; Performing space-time correlation on the characteristic data, the aging rate and the residual life evaluation value of the pipeline and the geometric mapping model, and establishing a digital twin physical layer model for simulating pipeline network pressure distribution, flow distribution and leakage physical process through a hydraulic calculation engine and a pipeline stress simulation model; Training a voiceprint large model and a time sequence prediction model based on historical leakage event data on the basis of the digital twin physical layer model, and constructing a digital twin behavior layer model for simulating leakage dynamic evolution behavior; based on the digital twin behavior layer model, taking a simulation result of the digital twin behavior layer model as an input to train a multi-objective optimization algorithm to output a final loss control strategy so as to obtain a digital twin decision layer model; And jointly outputting the multi-scale coupled digital twin model by combining the geometric mapping model, the digital twin behavior layer model, the digital twin physical layer model and the digital twin decision layer model.
3. The method for controlling leakage damage of a water supply network according to claim 1, wherein the identifying the leakage state in the water supply network based on the multi-scale coupled digital twin model and the fusion data thereof, and outputting the leakage evolution trend and the risk level of the leakage state through a time sequence prediction model comprises: Extracting voiceprint energy values, pressure fluctuation characteristic values and flow values of key nodes in current and historical periods of time from the multi-scale coupled digital twin model; Inputting the extracted voiceprint energy value, the pressure fluctuation characteristic value and the flow value of the key node into a pre-trained voiceprint large model, outputting the position information of the leakage point, and judging the leakage type; calculating the pipe network health index of the leakage point by combining the position information of the leakage point, the pipe material of the pipe section, the pipe age and the residual life evaluation value; And inputting the position information of the leakage points, the leakage type, the historical leakage change data and the pipe network health index into the time sequence prediction model, outputting a leakage increase prediction curve of a future designated period, and classifying risk grades according to the degree that the predicted leakage exceeds a preset threshold.
4. The method for controlling leakage and damage of a water supply pipe network according to claim 3, wherein the inputting the position information of the leakage point, the leakage type, the historical leakage change data and the pipe network health index into the time sequence prediction model, outputting a leakage increase prediction curve of a future designated period, and classifying risk levels according to the degree that the predicted leakage exceeds a preset threshold value comprises: Arranging the historical drain change data in time sequence to construct an input time sequence for model learning; The pipe network health index is used as a correction factor and is input into the input end of the time sequence prediction model together with the position information, the leakage type and the input time sequence of the leakage point; And learning the long-term dependency relationship of the input time sequence through a memory unit in the time sequence prediction model, and outputting leakage prediction values of a plurality of continuous time points in the future by combining the adjustment influence of the correction factors on the evolution speed to form a leakage increase prediction curve.
5. The method for controlling leakage damage of a water supply pipe network according to claim 1, wherein the generating a final loss control strategy by solving an optimal solution by using a multi-objective optimization algorithm and performing simulation verification by using the digital twin model based on the leakage evolution trend and the risk level with water saving benefit, user side water pressure influence and operation cost as optimization targets comprises: Determining an optimization target priority according to the risk level, wherein if the risk level is high risk, the water pressure of the user is preferentially ensured, and if the risk level is low risk, the water saving benefit is preferentially ensured; taking the opening degree of each controllable valve in the water supply network as a decision variable, and constructing a multi-objective optimization mathematical model comprising a water-saving benefit maximization function, a user side water pressure influence minimization function and a valve operation cost minimization function; In the simulation environment of the multi-scale coupling digital twin model, a multi-objective optimization algorithm is called, the leakage evolution trend is deduced for multiple times, and the realization value of each objective function under different valve opening combination schemes is estimated; searching and determining a group of pareto optimal solution sets for achieving non-inferior balance in each optimization target from mass deduction results through a multi-target optimization algorithm, and selecting a final loss control strategy from the pareto optimal solution sets.
6. The method for controlling leakage damage of water supply network according to claim 3, wherein the step of issuing a control command to a valve execution unit of the water supply network based on the active damage control decision scheme, and simultaneously pushing maintenance task information to an operation and maintenance system comprises the steps of: analyzing the final loss control strategy to obtain an equipment identifier, a target opening value and a downstream booster pump station identifier and an operation instruction which need to be linked of a valve to be regulated; The equipment identifier, the target opening value, the downstream booster pump station identifier and the operation instruction to be linked are issued to a programmable logic controller and a remote terminal unit on site through an industrial control protocol; based on the position information of the leakage points, the leakage type and the risk level, generating structured maintenance work order data comprising problem description, geographical position coordinates, suggested maintenance time priority and required spare part tools; Pushing the structured maintenance work order data to an intelligent terminal of an operation and maintenance person through an interface of a work order management system.
7. The method for controlling leakage damage of water supply network according to claim 2, wherein the step of issuing a control command to a valve execution unit of the water supply network based on the final damage control strategy and pushing maintenance task information to an operation and maintenance system comprises: acquiring actual leakage data of the leakage point again in a preset time period after the final loss control strategy is executed; Comparing the actual leakage data with the predicted value of the corresponding time point in the leakage increase prediction curve, and calculating the relative error; judging whether the relative error exceeds a preset allowable error threshold, if so, sorting a full-flow data set from feature extraction to effect verification of the leakage event into a new training sample; And performing iterative fine tuning on internal weight parameters of the time sequence prediction model by using the new training sample, and calibrating corrosion rate weight factors influencing prediction in the digital twin body behavior layer model.
8. The utility model provides a delivery pipe net leaks damage management and control device which characterized in that includes: The data acquisition and preprocessing module is used for acquiring acoustic signals, water pressure data and flow data of the water supply network, and extracting characteristics of the acoustic signals, the water pressure data and the flow data to obtain characteristic data representing the running state of the network; The digital twin modeling and fusion module is used for constructing a multi-scale coupling digital twin model capable of dynamically mapping and simulating the running state of the pipe network through space-time correlation fusion based on the spatial structure data, the pipe material, the pipe age historical data and the soil corrosion environment parameters of the water supply pipe network and the characteristic data, and obtaining corresponding fusion data; The leakage identification and evolution prediction module is used for identifying the leakage state in the water supply network based on the multi-scale coupling digital twin model and the fusion data thereof, and outputting the leakage evolution trend and the risk level of the leakage state through the time sequence prediction model; the multi-objective loss control decision module is used for solving an optimal solution by using a multi-objective optimization algorithm based on the leakage evolution trend and the risk level and taking water saving benefits, the influence of water pressure at a user side and operation cost as optimization targets, and performing simulation verification by using the multi-scale coupling digital twin model to generate a final loss control strategy; and the instruction execution and task dispatch module is used for issuing a control instruction to a valve execution unit of the water supply network based on the final loss control strategy and pushing maintenance task information to the operation and maintenance system.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements a water supply network leakage control method according to any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements a water supply pipe network leakage control method according to any one of claims 1 to 7.

Description

Method, device, equipment and medium for controlling leakage damage of water supply pipe network Technical Field The invention relates to the technical field of water supply network leakage damage control, in particular to a water supply network leakage damage control method, a device, equipment and a medium. Background The water supply network is an important component of urban lifeline engineering. The pipe network leakage not only causes huge water resource waste and economic loss, but also can cause secondary disasters such as pavement collapse, water quality pollution and the like. Therefore, finding and controlling the leakage of the pipe network in time is one of the core targets of intelligent water affair construction. Currently, the mainstream pipe network leakage monitoring technology mainly comprises statistical analysis based on water pressure and flow data and fixed-point or mobile type hearing leakage detection based on acoustic signals. In addition, the acoustic leakage is highly dependent on manual experience, is greatly interfered by environmental noise, and is difficult to accurately position leakage points and judge evolution trend, so that the best control and loss time is missed, the tiny leakage finally develops into a pipe explosion, and the water resource waste is increased. Disclosure of Invention The invention aims to provide a method, a device, equipment and a medium for controlling leakage and damage of a water supply network, which are used for identifying the leakage and damage state of the water supply network, actively predicting evolution trend and generating an optimal solution and control damage strategy so as to reduce the damage degree of the leakage and damage of the water supply network. The invention provides a water supply pipe network leakage damage management and control method, which comprises the steps of obtaining acoustic signals, water pressure data and flow data of a water supply pipe network, carrying out feature extraction on the acoustic signals, the water pressure data and the flow data to obtain feature data representing the running state of the pipe network, based on spatial structure data, pipeline materials, pipe age historical data and soil corrosion environment parameters of the water supply pipe network and the feature data, constructing a multi-scale coupling digital twin model capable of dynamically mapping and simulating the running state of the pipe network through time-space correlation fusion, obtaining corresponding fusion data, based on the multi-scale coupling digital twin model and the fusion data, identifying the leakage damage state in the water supply pipe network, outputting the leakage damage evolution trend and risk level of the leakage damage state through a time sequence prediction model, solving the optimal solution through a multi-scale coupling digital twin model based on the space structure data, pipeline materials, pipe age historical data and the soil corrosion environment parameters of the water supply pipe network, generating a final control strategy, and simultaneously sending a task maintenance instruction to a water supply pipe network maintenance control unit based on the final control strategy, and carrying out task maintenance system maintenance instruction. According to one embodiment of the invention, the method comprises the steps of constructing a multi-scale coupling digital twin model capable of dynamically mapping and simulating the running state of a pipe network through space-time correlation fusion based on spatial structure data, pipe materials, pipe age historical data and soil corrosion environment parameters of the water supply pipe network and the characteristic data, and obtaining corresponding fusion data, wherein the method comprises the steps of calling a design drawing of the water supply pipe network and geographic information system data, and constructing a geometric mapping model containing geometric shapes and spatial position information of pipes, valves and pump stations in a three-dimensional graphic engine; obtaining historical data of pipeline material parameters, pipeline wall thickness and pipeline age of the water supply network, combining the soil corrosion environment parameters, calculating an ageing rate and a residual life evaluation value of a pipeline through a material corrosion dynamics model, performing time-space correlation on the characteristic data, the ageing rate and the residual life evaluation value of the pipeline and the geometric mapping model, establishing a digital twin physical layer model for simulating pipeline pressure distribution, flow distribution and leakage physical process through a hydraulic calculation engine and a pipeline stress simulation model, training a voiceprint large model and a time sequence prediction model based on historical leakage event data on the digital twin physical layer model, constructing a digital twin behavior layer model for