| CPC G01M 3/04 (2013.01) | 17 Claims |
|
1. A method for gas monitoring and assessment based on Internet of Things, wherein the method is executed by a processor in a smart gas safety management platform of a system for gas monitoring and assessment based on Internet of Things, and the method comprises:
obtaining industrial gas data of a workshop to be monitored based on a smart gas indoor device object platform and uploading the industrial gas data to a smart gas data center through a smart gas indoor device sensor network platform;
obtaining regional data of the workshop to be monitored based on a smart gas service platform and uploading the regional data to the smart gas data center through the smart gas indoor device sensor network platform;
determining a key sub-region based on the regional data and the industrial gas data;
determining at least one set of diffusion trend data of the key sub-region based on environmental data of the key sub-region;
determining at least one recommended monitoring site based on the at least one set of diffusion trend data of the key sub-region, wherein the at least one recommended monitoring site includes at least one mandatory turn-on site and at least one optional turn-on site;
sending the at least one recommended monitoring site to a smart gas user platform through the smart gas indoor device sensor network platform, and integrating and coordinating an interaction between the smart gas user platform and a smart supervision service sub-platform to obtain service for safety supervision;
constructing a map of recommended monitoring sites based on the at least one recommended monitoring site, the map of recommended monitoring sites being a map that reflects a relationship between various factors in symbolic form based on the at least one recommended monitoring site; wherein the map of recommended monitoring sites includes nodes and edges; the nodes refer to the recommended monitoring sites, and attributes of the node include a height of a recommended monitoring site corresponding to the node and industrial gas-related data, the height refers to a rising distance of the recommended monitoring site in a vertical direction based on a ground of the workshop to be monitored, the industrial gas-related data refers to a situation of distribution and usage of gas pipelines and gas devices near the recommended monitoring site; the edges are used to connect the nodes, in response to that a distance between recommended monitoring sites corresponding to two nodes is less than a distance threshold, the two nodes are connected by an edge; the edge connecting nodes is a directed edge, and a direction of the edge is a diffusion direction of gas, attributes of the edge include the distance between recommended monitoring sites corresponding to the nodes and the diffusion direction of gas; and
determining a risk value through a risk assessment model based on the map of recommended monitoring sites, wherein the risk assessment model is a graph neural network model, and the risk value is an index used to judge whether the gas leakage is monitored timely and accurately; wherein the recommended monitoring sites contained in the map of recommended monitoring sites are all regarded as turn-on sites, an input of the risk assessment model includes the map of recommended monitoring sites, an output of the risk assessment model includes the risk value of a set of the turn-on sites included in the map of recommended monitoring sites; the smaller the risk value, the timelier and more accurate monitoring of the gas leakage; wherein
the risk assessment model is obtained by training with a training sample and a label separately, the training sample and the label are constructed with historical data of the smart gas data center,
the training sample includes at least one map of sample recommended monitoring site constructed based on different sample recommended monitoring sites, the label corresponding to the training sample is whether gas leakage is detected timely and accurately when monitoring devices are turned on according to recommended monitoring sites contained in the map of sample recommended monitoring site, and in response to determining that the gas leakage is detected, the label is 1, otherwise, the label is 0.
|