Large model-based systems and methods of internet of things (iot) for emergency regulation of water supply pipeline networks in smart cities

Patent metadata

Field	Value
Publication number	US-2026089217-A1
Application number	US-202519404075-A
Country	US
Kind code	A1
Filing date	Dec 1, 2025
Priority date	Jul 28, 2025
Publication date	Mar 26, 2026
Grant date	—

Abstract

Provide are a large model-based system and method of Internet of Things (IoT) for emergency regulation of a water supply pipeline network in a smart city. The system comprises a government regulation management platform, a government regulation sensing network platform, a government regulation object platform, a water company sensing network platform, and a smart water device object platform. The government regulation management platform is configured to: determine a time-series flow rate corresponding to each of a plurality of water pipeline node groups; determine a water and soil loss coefficient corresponding to each of one or more target regions based on the time-series flow rate; generate, based on the one or more water and soil loss coefficients, a temporary control parameter through a parameter generation model; and control an opening level of a valve corresponding to the one or more target regions based on the temporary control parameter.

First claim

What is claimed is:

1 . A large model-based system of internet of things (IoT) for emergency regulation of a water supply pipeline network in a smart city, comprising a government regulation management platform, a government regulation sensing network platform, a government regulation object platform, a water company sensing network platform, a smart water device object platform, wherein
the government regulation object platform includes a water company management platform; the smart water device object platform includes a monitoring device and at least one adjustment device; the government regulation management platform is configured to:
determine a time-series flow rate corresponding to each of a plurality of water pipeline node groups based on water flow features of a plurality of water pipeline nodes in the water supply pipeline network;
determine one or more water and soil loss coefficients corresponding to each of one or more target regions based on the time-series flow rates;
generate, based on the one or more water and soil loss coefficients, a temporary control parameter through a parameter generation model, the parameter generation model being a machine learning model; and
control an opening level of each of one or more valves corresponding to the one or more target regions based on the temporary control parameter.

2 . The system of claim 1 , wherein the smart water device object platform further includes a detection robot, and the government regulation management platform is further configured to:
collect a soil feature of each of the water pipeline node groups in each of the one or more target regions via the detection robot; construct a water and soil loss map based on the soil features and the time-series flow rates, wherein the water and soil loss map includes a plurality of nodes and a plurality of edges; and determine the one or more water and soil loss coefficients through a coefficient generation model based on the water and soil loss map, the coefficient generation model being a machine learning model.

3 . The system of claim 2 , wherein a node feature of each of the plurality of nodes includes one or more node-hotspot offsets of the node, and each of the one or more node hotspot offsets is a distance between the node and a target control hotspot.

4 . The system of claim 2 , wherein a node feature of each of the plurality of nodes includes a detection parameter of the detection robot.

5 . The system of claim 2 , wherein the water and soil loss map includes at least one key edge, and an edge feature of the at least one key edge includes a loss correlation coefficient.

6 . The system of claim 5 , wherein each of the at least one key edge is determined based on a time-series flow rate difference between two nodes connected by the edge.

7 . The system of claim 1 , wherein the government regulation management platform is further configured to:
determine at least one control hotspot based on a pipeline network pressure map and the one or more water and soil loss coefficients; and generate the temporary control parameter based on the at least one control hotspot.

8 . The system of claim 7 , wherein the government regulation management platform is further configured to:
determine one or more highly sensitive regions in the pipeline network pressure map based on a plurality of historical pressure maps and a plurality of historical loss coefficients in a predetermined historical time period; and determine the at least one control hotspot based on the one or more highly sensitive regions and the one or more water and soil loss coefficients.

9 . The system of claim 1 , wherein the temporary control parameter includes output power and an on/off parameter, and the government regulation management platform is further configured to:
control an operation of a water supply pump based on the output power; and control an operation of a user device based on the on/off parameter.

10 . A method for emergency regulation of a water supply pipeline network in a smart city, executed by a government regulation management platform of a large model based system of internet of things (IoT) for emergency regulation of the water supply pipeline network in the smart city, the system comprising a government regulation management platform, a government regulation sensing network platform, a government regulation object platform, a water company sensing network platform, a smart water device object platform, wherein
the government regulation object platform includes a water company management platform; the smart water device object platform includes a monitoring device and at least one adjustment device; the method comprising: determining a time-series flow rate corresponding to each of a plurality of water pipeline node groups based on water flow features of a plurality of water pipeline nodes in the water supply pipeline network; determining one or more water and soil loss coefficients corresponding to each of one or more target regions based on the time-series flow rates; generating, based on the one or more water and soil loss coefficients, a temporary control parameter through a parameter generation model, the parameter generation model being a machine learning model; and controlling an opening level of each of one or more valves corresponding to the one or more target regions based on the temporary control parameter.

11 . The method of claim 10 , wherein the smart water device object platform further includes a detection robot, the determining one or more soil erosion coefficients corresponding to each of one or more target regions based on the time-series flow rates includes:
collecting a soil feature of each of the water pipeline node groups in each of the one or more target regions via the detection robot; constructing a water and soil loss map based on the soil features and the time-series flow rates, wherein the water and soil loss map includes a plurality of nodes and a plurality of edges; and determining the one or more water and soil loss coefficients through a coefficient generation model based on the water and soil loss map, the coefficient generation model being a machine learning model.

12 . The method of claim 11 , wherein a node feature of each of the plurality of nodes includes a node-hotspot offset of the node, and the node hotspot offset is a distance between the node and a target control hotspot.

13 . The method of claim 11 , wherein a node feature of each of the plurality of nodes includes a detection parameter of the detection robot.

14 . The method of claim 11 , wherein the water and soil loss map includes at least one key edge, and an edge feature of the at least one key edge includes a loss correlation coefficient.

15 . The method of claim 14 , wherein each of the at least one key edge is determined based on a time-series flow rate difference between two nodes connected to the edge.

16 . The method of claim 10 , further comprising:
determining at least one control hotspot based on a pipeline network pressure map and the water and soil loss coefficient; and generating the temporary control parameter based on the at least one control hotspot.

17 . The method of claim 16 , further comprising:
determining one or more highly sensitive regions in the pipeline network pressure map based on a plurality of historical pressure maps and a plurality of historical loss coefficients in a predetermined historical time period; and determining the at least one control hotspot based on the one or more highly sen

Assignees

• Chengdu Qinchuan Iot Tech Co Ltd

Inventors

• Shao Hanshu

Classifications

• G06Q50/26

  Government or public services (business processes related to the transportation industry G06Q50/40) · CPC title

• G06Q50/06

  Energy or water supply · CPC title

• G16Y20/30

  relating to resources, e.g. consumed power · CPC title

• G16Y10/35

  Utilities, e.g. electricity, gas or water · CPC title

• G16Y40/20

  Analytics; Diagnosis · CPC title