acquiring gas attribute data of gas passing through a metering device based on a distributed sensor deployed in a gas pipeline network, wherein the gas attribute data includes gas density, composition and mass, fluid pulsation, and flow rate profile;

determining a first accuracy risk of the metering device through performing pre-diagnosis on the metering device based on the gas attribute data and metering device information, wherein the metering device information includes a type, a service life, and a location, and the first accuracy risk refers to a pre-diagnosed possible abnormal metering device and an abnormal possibility thereof;

in response to the first accuracy risk of at least one metering device satisfying a preset risk condition, constructing an accuracy diagnosis map based on metering devices in a gas area, wherein nodes of the accuracy diagnosis map include the metering devices, edges of the accuracy diagnosis map include gas pipelines between the nodes, attributes of the nodes include types of the metering devices, locations of the nodes in the gas pipeline network, gas monitoring data, gas theoretical data, the first accuracy risk, flow rate monitoring data, and pressure monitoring data;

determining, based on the accuracy diagnosis map, a second accuracy risk of the at least one metering device in the gas area through a pipeline network diagnosis model, the pipeline network diagnosis model being a machine learning model; wherein the second accuracy risk refers to a possibility of accuracy abnormality of the possible abnormal metering device, and the preset risk condition refers to a count of metering devices with the first accuracy risk exceeding a first risk threshold reaching a quantity alert value; and

displaying detection indication information to the smart gas device management platform based on the second accuracy risk and a risk threshold.