Drainage pipe network sensing equipment deployment optimization method

By employing dynamic graph modeling and multi-objective optimization methods, combined with incremental learning mechanisms and an improved NSGA-III algorithm, the problem of low computational efficiency in the deployment of sensors in drainage pipe networks was solved, achieving adaptive monitoring and reducing the cost of equipment.

CN121543419APending Publication Date: 2026-02-17HARBIN AEROSPACE STAR DATA SYST TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511709868.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-20
Publication Date
2026-02-17

AI Technical Summary

Technical Problem

Existing technologies cannot dynamically adapt to sudden changes in flow and topology modifications in drainage pipe networks, resulting in low computational efficiency in sensor deployment and difficulty in coordinating multi-objective optimization, making it impossible to balance coverage, cost, and data quality.

Method used

Employing dynamic graph modeling, multi-objective optimization, and an improved NSGA-III algorithm, combined with an incremental learning mechanism, this approach optimizes equipment deployment by constructing a time-series graph structure and multi-objective functions. It supports real-time adjustments and local search strategies to address network changes.

Benefits of technology

It enables adaptive quality monitoring under dynamic changes in the pipeline network, reduces equipment deployment and maintenance costs, and improves computing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121543419A_ABST
    Figure CN121543419A_ABST
Patent Text Reader

Abstract

The invention relates to a drainage pipe network sensing equipment deployment optimization method, and belongs to the technical field of intelligent water affairs. The method comprises the steps that firstly, a drainage pipe network is abstracted into a time sequence diagram structure through dynamic diagram modeling, node attributes comprise time sequence data such as liquid level and flow, and edge attributes are introduced into propagation time delay weights reflecting dynamic hydraulic correlation; on the basis, an optimization model with the purposes of minimizing the total cost of equipment and maximizing the coverage information entropy and the redundancy coverage degree is established. In order to solve the problem of dynamic adaptability, the model integrates an incremental learning mechanism, when it is detected that a pipe network topology change or an attribute exceeds a threshold value, a local optimization process is automatically triggered, an improved NSGA-III algorithm is adopted to efficiently solve an affected sub-graph, and a Pareto optimal point distribution scheme is rapidly generated. According to the method, the limitation of a traditional static deployment method is overcome, the method can adapt to dynamic changes of the pipe network, and the economical efficiency and the calculation efficiency of the scheme are remarkably improved while the monitoring quality is guaranteed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a method for optimizing the deployment of sensing devices in drainage pipe networks, belonging to the field of smart water management technology. Background Technology

[0002] Deploying monitoring equipment in drainage pipe networks is the most direct way to continuously monitor the operational status of drainage pipe networks and to diagnose faults and provide early warnings of anomalies in urban drainage systems. Traditional methods rely on expert experience to deploy sensors in a fixed manner, which cannot adapt to dynamic scenarios such as sudden changes in pipe network flow and topology modifications. Existing technologies lack adaptability to dynamic pipe network topologies, and the optimization model cannot be locally updated, resulting in low computational efficiency. Existing technical solutions focus on single-objective optimization and fail to coordinate the conflicts between coverage, cost, and data quality. Summary of the Invention

[0003] This invention aims to solve the problems of the inability to dynamically adapt, difficulty in balancing multiple objectives, and low computational efficiency in the deployment of sensors in drainage pipe networks, and proposes an optimization method for the deployment of sensing devices in drainage pipe networks.

[0004] The technical solution of the present invention:

[0005] A method for optimizing the deployment of sensing devices in a drainage pipe network includes the following steps:

[0006] S1. Dynamic graph modeling: The drainage network is abstracted into a time-series graph structure; S2. Multi-objective optimization: A multi-objective optimization model is constructed, integrating an incremental learning mechanism; S3. Dynamic optimization solution: An improved NSGA-III algorithm is used to solve the dynamic Pareto optimal solution set, supporting manual intervention and real-time adjustment.

[0007] Specifically, step S1 is as follows:

[0008] S1.1 Construct the sequence diagram structure, Among them, node attributes include real-time liquid level. Traffic time series data Pipe diameter and slope are encoded into a temporal feature vector using LSTM; edge attributes are used to introduce hydraulic propagation delay weights. The calculation formula is:

[0009]

[0010] Corr() is the correlation coefficient of the flow time series data of adjacent nodes, reflecting the dynamic hydraulic correlation. For time-varying flow velocity, This is a weighting adjustment factor;

[0011] S1.2 Update node and edge attributes using a sliding window, supporting real-time data stream input.

[0012] Specifically, step S2 includes:

[0013] S2.1. Constructing a multi-objective function: ,in Indicates the total cost of the equipment. For equipment installation costs, Indicates whether to deploy points; This represents the entropy of information coverage, maximizing the diversity of information coverage at key nodes; Node v is covered by at least k monitoring points to ensure redundant monitoring of critical nodes;

[0014] S2.2. Set coverage radius constraints: ,in R is the average flow rate, and R is the maximum response time.

[0015] S2.3. When a change in the pipeline topology is detected or Incremental optimization is initiated at that time.

[0016] Specifically, step S3 employs an improved NSGA-III algorithm, introduces a local search strategy, generates an initial solution based on the historical Pareto front, recalculates the importance of nodes in the subgraph affected by network changes, optimizes only the node placement scheme within the subgraph, and reduces computational complexity.

[0017] The beneficial effects of this invention are:

[0018] This invention can adapt to dynamic changes in the pipeline network while ensuring monitoring quality, significantly reducing equipment deployment and maintenance costs, and greatly improving optimization calculation efficiency. Attached Figure Description

[0019] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0020] Example 1:

[0021] like Figure 1 As shown, a method for optimizing the deployment of sensing devices in a drainage pipe network includes the following steps:

[0022] S1. Dynamic Graph Modeling: Abstracting the drainage network into a time-series graph structure, and constructing the time-series graph structure. Among them, node attributes include real-time liquid level. Traffic time series data Pipe diameter ,slope Encode into temporal feature vectors using LSTM. ℝ d Edge attributes introduce hydraulic propagation delay weights. The calculation formula is:

[0023]

[0024] Corr() is the correlation coefficient of the flow time series data of adjacent nodes, reflecting the dynamic hydraulic correlation. The time-varying velocity is calculated based on Manning's formula. The weighting factor balances flow correlation and physical propagation characteristics. Edge weights are updated using a sliding window, each... The system updates node and edge attributes at time steps, supporting real-time data stream input. S2. Multi-objective optimization: Constructing a multi-objective optimization model and integrating an incremental learning mechanism;

[0025] S2.1. Constructing a multi-objective function: ,in Indicates the total cost of the equipment. For equipment installation costs, Indicates whether to deploy points; This represents the entropy of information coverage, maximizing the diversity of information coverage at key nodes; Node v is covered by at least k monitoring points to ensure redundant monitoring of critical nodes;

[0026] S2.2. Set coverage radius constraints: ,in R is the average flow rate, and R is the maximum response time.

[0027] S2.3. When a change in the pipeline topology is detected or When the time comes, incremental optimization is initiated. S3. Dynamic optimization solution: An improved NSGA-III algorithm is adopted, introducing a local search strategy. An initial solution is generated based on the historical Pareto front. The importance of nodes in the subgraph affected by pipeline changes is recalculated. Only the node placement scheme within the subgraph is optimized to reduce computational complexity.

Claims

1. A method for optimizing the deployment of sensing devices in a drainage pipe network, characterized in that, Includes the following steps: S1. Dynamic graph modeling: The drainage network is abstracted into a time-series graph structure; S2. Multi-objective optimization: A multi-objective optimization model is constructed, integrating an incremental learning mechanism; S3. Dynamic optimization solution: An improved NSGA-III algorithm is used to solve the dynamic Pareto optimal solution set, supporting manual intervention and real-time adjustment.

2. The method for optimizing the deployment of drainage pipe network sensing equipment according to claim 1, characterized in that, In step S1, step S1 specifically includes: S1.1 Construct the sequence diagram structure, Among them, node attributes include real-time liquid level. Traffic time series data Pipe diameter and slope are encoded into a temporal feature vector using LSTM; edge attributes are used to introduce hydraulic propagation delay weights. The calculation formula is: Corr() is the correlation coefficient of the flow time series data of adjacent nodes, reflecting the dynamic hydraulic correlation. For time-varying flow velocity, This is a weighting adjustment factor; S1.2 Update node and edge attributes using a sliding window, supporting real-time data stream input.

3. The method for optimizing the deployment of drainage pipe network sensing equipment according to claim 1, characterized in that, Step S2 specifically includes: S2.

1. Constructing a multi-objective function: ,in Indicates the total cost of the equipment. For equipment installation costs, Indicates whether to deploy points; This represents the entropy of information coverage, maximizing the diversity of information coverage at key nodes; Node v is covered by at least k monitoring points to ensure redundant monitoring of critical nodes; S2.

2. Set coverage radius constraints: ,in R is the average flow rate, and R is the maximum response time. S2.

3. When a change in the pipeline topology is detected or Incremental optimization is initiated at that time.

4. The method for optimizing the deployment of drainage pipe network sensing equipment according to claim 1, characterized in that, Step S3 employs an improved NSGA-III algorithm, introducing a local search strategy. It generates an initial solution based on the historical Pareto front, recalculates the importance of nodes in the subgraph affected by network changes, and optimizes only the node placement scheme within the subgraph, thereby reducing computational complexity.