Back to results
US 12,614,240 B2Granted
Method for smart gas pipeline network inspection and internet of things system thereof
AssigneeChengdu Qinchuan Iot Technology Co., Ltd.
InventorsZehua Shao, Yong Li, Lei Zhang
Life of this patent
Estimating the term…
Abstract
The embodiments of the present disclosure provide a method for smart gas pipeline network inspection, implemented on a smart gas pipeline network security management platform based on an Internet of Things system for smart gas pipeline network inspection, and the method comprising: obtaining a gas pipeline network distribution; determining at least one inspection sub-area based on the gas pipeline network distribution; determining, based on the at least one inspection sub-area, an inspection plan for each of the at least one inspection sub-area, the inspection plan at least including an inspection frequency.
Claims18
A method for smart gas pipeline network inspection, implemented on a smart gas pipeline network security management platform based on an Internet of Things system for smart gas pipeline network inspection, and the method comprising:
- obtaining a gas pipeline network distribution;
- determining at least one inspection sub-area based on the gas pipeline network distribution, including:
- determining at least one inspection personnel station based on the gas pipeline network distribution, including:
- constructing a first pipeline network graph based on the gas pipeline network distribution, wherein nodes of the first pipeline network graph correspond to pipeline network branches in the gas pipeline network distribution;
- edges of the first pipeline network graph correspond to pipelines in the gas pipeline network distribution;
- and each edge of the first pipeline network graph corresponds to a pipeline connecting two pipeline network branches;
- outputting, based on the nodes of the first pipeline network graph and/or the edges of the first pipeline network graph through a probability determination model, a probability that the nodes of the first pipeline network graph and/or the edges of the first pipeline network graph are inspection personnel stations, wherein
- the probability determination model is a trained graph neural network model, the probability determination model is obtained by training a plurality of sample pipeline network graphs with labels, and a label setting manner includes:
- in each of the plurality of sample pipeline network graphs, setting a label of a node or an edge that is actually set as a inspection personnel station to 1;
- and
- setting values of labels of other nodes or edges in a range of [0,1] based on a preset attenuation degree;
- a training process of the probability determination model includes:
- inputting each sample pipeline network graph into the probability determination model;
- obtaining a probability value of each node and edge as a inspection personnel station outputted by the probability determination model based on the each node and edge in the sample pipeline network graph;
- constructing a loss function based on a label of each sample pipeline network graph and probability values outputted by the probability determination model;
- and
- obtaining a trained probability determination model until preset conditions are satisfied, wherein the preset conditions include the loss function being less than a first threshold, convergence, or the training cycle reaching a second threshold;
- and
- determining the at least one inspection personnel station based on an output of the nodes of the first pipeline network graph and the edges of the first pipeline network graph;
- and
- determining the at least one inspection sub-area based on the gas pipeline network distribution and the at least one inspection personnel station;
- determining, based on the at least one inspection sub-area, an inspection plan for each of the at least one inspection sub-area, the inspection plan at least including an inspection frequency;
- and
- controlling at least one inspection engineering-related device to perform a corresponding inspection operation on a pipeline network device based on the inspection plan, wherein the at least one inspection engineering-related device includes an alarm device, and the pipeline network device includes the pipelines in the gas pipeline network distribution and gate stations.
21 more elements
An Internet of Things system for smart gas pipeline network inspection, comprising:
- a smart gas user platform, a smart gas service platform, a smart gas pipeline network security management platform, a smart gas sensor network platform, and a smart gas object platform;
- wherein
- the smart gas object platform is configured to obtain a gas pipeline network distribution, and transmit the gas pipeline network distribution to the smart gas pipeline network security management platform through the smart gas sensor network platform;
- the smart gas pipeline network security management platform is configured to:
- determine at least one inspection sub-area based on the gas pipeline network distribution;
- wherein to determine the at least one inspection sub-area based on the gas pipeline network distribution, the smart gas pipeline network security management platform is further configured to:
- determine at least one inspection personnel station based on the gas pipeline network distribution, wherein to determine the at least one inspection personnel station based on the gas pipeline network distribution, the smart gas pipeline network security management platform is further configured to:
- construct a first pipeline network graph based on the gas pipeline network distribution, wherein the nodes of the first pipeline network graph correspond to pipeline network branches in the gas pipeline network distribution;
- edges of the first pipeline network graph correspond to pipelines in the gas pipeline network distribution;
- and each edge of the first pipeline network graph corresponds to a pipeline connecting two pipeline network branches;
- output, based on the nodes of the first pipeline network graph and/or the edges of the first pipeline network graph through a probability determination model, a probability that the nodes of the first pipeline network graph and/or the edges of the first pipeline network graph are inspection personnel stations, wherein
- the probability determination model is a trained graph neural network model, the probability determination model is obtained by training a plurality of sample pipeline network graphs with labels, and a label setting manner includes:
- in each of the plurality of sample pipeline network graphs, setting a label of a node or an edge that is actually set as a inspection personnel station to 1;
- and
- setting values of labels of other nodes or edges in a range of [0,1], based on a preset attenuation degree;
- a training process of the probability determination model includes:
- inputting each sample pipeline network graph into the probability determination model;
- obtaining a probability value of each node and edge as a inspection personnel station outputted by the probability determination model based on the each node and edge in the sample pipeline network graph;
- constructing a loss function based on a label of each sample pipeline network graph and probability values outputted by the probability determination model;
- and
- obtaining a trained probability determination model until preset conditions are satisfied, wherein the preset conditions include the loss function being less than a first threshold, convergence, or the training cycle reaching a second threshold;
- and
- determine the at least one inspection personnel station based on an output of the nodes of the first pipeline network graph and the edges of the first pipeline network graph;
- and
- determine the at least one inspection sub-area based on the gas pipeline network distribution and the at least one inspection personnel station;
- determine, based on the at least one inspection sub-area, an inspection plan for each of the at least one inspection sub-area, the inspection plan at least including an inspection frequency;
- and
- control at least one inspection engineering-related device to perform a corresponding inspection operation on a pipeline network device based on the inspection plan, wherein
- the at least one inspection engineering-related device includes an alarm device, and
- the pipeline network device includes the pipelines in the gas pipeline network distribution and gate stations;
- the smart gas service platform is configured to feed back the inspection plan to the smart gas user platform.
28 more elements
Dependencies are inferred from the claim text.