An Internet of Things (IoT) system for corrosion protection optimization of a pipeline of smart gas, wherein the IoT system comprises a smart gas safety management platform, a smart gas user platform, a smart gas service platform, a smart gas pipeline network equipment sensor network platform, and a smart gas pipeline network equipment object platform, the smart gas safety management platform includes a smart gas safety management sub-platform and a smart gas data center, the smart gas pipeline network equipment object platform is configured to obtain inspection data, and transmit the inspection to the smart gas safety management platform through the smart gas pipeline network equipment sensor network platform, the smart gas user platform includes a gas user sub-platform and a supervision user sub-platform, the smart gas service platform includes a smart gas usage service sub-platform and a smart supervision service sub-platform;
- and
- the smart gas safety management platform is configured to:
- obtain inspection data of a gas pipeline in a gas pipeline network, the inspection data including gas monitoring data and in-depth inspection data of the gas pipeline, wherein the in-depth inspection data of the gas pipeline refers to data obtained by performing in-depth inspection on the estimated pipeline corrosion regions, the in-depth inspection data of the gas pipeline includes thickness inspection data and image inspection data, the thickness detection data refers to a thickness of the gas pipeline in the estimated pipeline corrosion regions, and the image detection data refers to an image of the inside of the gas pipeline of the estimated pipeline corrosion regions;
- construct a first pipeline diagram of the gas pipeline based on the inspection data, wherein the first pipeline diagram reflects a connection relationship between first pipelines, the first pipeline diagram includes nodes and edges, the nodes in the first pipeline diagram include a first type of nodes corresponding to monitoring device installation points and a second type of nodes corresponding to pipeline demarcation points, features of the first type of nodes include the inspection data, features of the second type of nodes is null, edge features in the first pipeline diagram include a gas flow direction, pipeline features, and an environmental unit feature sequence, the pipeline features refer to features related to the properties of the gas pipeline, the pipeline features include a length of a pipeline connecting two nodes, an inner diameter of a pipeline, and a pipe wall material, the environmental unit features are represented by the environmental unit feature sequence, and the environmental unit features refer to features of an environment where a unit pipeline between any two nodes in the first pipeline diagram is located;
- determine a corrosion probability of the gas pipeline through a corrosion probability prediction model based on the first pipeline diagram, wherein corrosion probability prediction model is a graph neural network model (GNN), an input of the corrosion probability prediction model is the first pipeline diagram, and an output of the corrosion probability prediction model is a corrosion probability of each edge of the first pipeline diagram;
- determine one or more estimated pipeline corrosion regions based on the corrosion probability of the gas pipeline;
- obtain the in-depth inspection data of the gas pipeline by performing in-depth inspection on at least one of the estimated pipeline corrosion regions;
- determine corrosion features of the one or more estimated pipeline corrosion regions based on the gas monitoring data and in-depth inspection data;
- wherein the corrosion features include a corrosion type and a corrosion degree, the estimated pipeline corrosion regions include a first estimated corrosion region and a second estimated corrosion region, the first estimated corrosion region refers to an estimated pipeline corrosion region that is detected by an instrument to obtain the in-depth detection data of the gas pipeline, the second estimated corrosion region refers to an estimated pipeline corrosion region that fails to be detected by the instrument to obtain the in-depth inspection data of the gas pipeline, and the determine corrosion features of the one or more estimated pipeline corrosion regions includes:
- determining a corrosion type and a corrosion degree of the first estimated corrosion region based on the in-depth inspection data of the gas pipeline of the first estimated corrosion region;
- and
- determining a corrosion type and a corrosion degree of the second estimated corrosion region through the corrosion feature prediction model based on a second pipeline diagram of the gas pipeline network;
- wherein the corrosion feature prediction model is the GNN, an input of the corrosion feature prediction model is the second pipeline diagram, an output of the corrosion feature prediction model is the corrosion type and a corrosion degree, the second pipeline diagram includes nodes and edges, the nodes in the second pipeline diagram reflect locations where gas monitoring devices are installed and/or locations where gas monitoring devices are not installed, the edges in the second pipeline diagram reflect the gas pipelines, the edges in the second pipeline diagram include a first type of edges and a second type of edges, the first type of edges represent the gas pipelines of the first estimated corrosion region, and the second type of edges represent the gas pipelines of the second estimated corrosion region;
- wherein node features of the second pipeline diagram include at least the monitoring data of the gas monitoring devices, edge features of the second pipeline diagram include at least gas flow directions of the gas pipelines, pipeline features, environmental features, and the corrosion type and the corrosion degree of the first estimated corrosion region, the corrosion type and the corrosion degree are set according to an actual situation for the first type of edges, and the corrosion type and the corrosion degree are set to a preset value for the second type of edges;
- the edge features further include an inherent confidence of the gas pipeline, the inherent confidence is determined based on a confidence prediction model, the confidence prediction model is a machine learning model, an input of the confidence prediction model includes a distance between the first estimated corrosion region and the current second estimated corrosion region, a gas pipeline complexity, and a gas monitoring data saturation, the gas monitoring data saturation is determined based on a count of gas monitoring nodes and a degree of dispersion of the gas monitoring nodes in the gas pipeline network, and the corrosion feature prediction model is further used to update the inherent confidence to obtain a predicted confidence for the second estimated corrosion region;
- and predict the corrosion features of the second estimated corrosion region through the in-depth detection data of the gas pipeline of the first estimated corrosion region based on the second pipeline diagram;
- and
- determine a repair plan based on a corrosion situation of a pipe wall.