US 12,072,066 B2
Methods and internet of things (IoT) systems for gas loss control based on smart gas platform
Zehua Shao, Chengdu (CN); Yong Li, Chengdu (CN); and Junyan Zhou, Chengdu (CN)
Assigned to CHENGDU QINCHUAN IOT TECHNOLOGY CO., LTD., Chengdu (CN)
Filed by CHENGDU QINCHUAN IOT TECHNOLOGY CO., LTD., Sichuan (CN)
Filed on Nov. 23, 2023, as Appl. No. 18/518,478.
Claims priority of application No. 202311396477.1 (CN), filed on Oct. 26, 2023.
Prior Publication US 2024/0084976 A1, Mar. 14, 2024
Int. Cl. G16Y 40/10 (2020.01); F17D 5/00 (2006.01); G16Y 10/35 (2020.01); G16Y 20/30 (2020.01); G16Y 40/35 (2020.01)
CPC F17D 5/005 (2013.01) [G16Y 10/35 (2020.01); G16Y 20/30 (2020.01); G16Y 40/10 (2020.01); G16Y 40/35 (2020.01)] 4 Claims
OG exemplary drawing
 
1. A method for gas loss control based on a smart gas platform executed by a smart gas device management platform of an Internet of Things (IoT) system for gas loss control based on a smart gas platform, comprising:
obtaining gas delivery parameters of a preset point in a gas pipeline branch, wherein the gas delivery parameters include a pressure, a temperature, and a gas flow rate of gas delivery;
obtaining gas data corresponding to the gas pipeline branch;
determining a gas loss rate of the gas pipeline branch based on the gas delivery parameters and the gas data, wherein the gas data is obtained based on readings of a gas metering device, the gas data includes at least a total amount of gas supply and a total amount of gas consumption; and
determining the gas loss rate of the gas pipeline branch includes:
determining an estimated gas loss rate of the gas pipeline branch based on the total amount of gas supply and the total amount of gas consumption;
determining a confidence level of the estimated gas loss rate by a first prediction model based on metering device data and the gas delivery parameters, the first prediction model being a machine learning model; and
constructing a sequence of confidence levels based on the confidence level;
constructing a to-be-matched vector based on the sequence of confidence levels, searching in a database based on the to-be-matched vector, determining a history vector that meets a matching condition as a target vector; and
determining a reference gas loss rate corresponding to the target vector as the gas loss rate;
in response to the gas loss rate satisfying a first predetermined condition, generating at least one candidate operating parameter;
predicting a predicted loss rate corresponding to each of the at least one candidate operating parameter based on a second prediction model, wherein the second prediction model is a machine learning model;
determining a target operating parameter based on a candidate operating parameter with a predicted loss rate satisfying a second predetermined condition in the at least one candidate operating parameter; and
adjusting a pressure of a gas pipeline network based on the target operating parameter.