NATURAL GAS DEMAND PREDICTING METHOD USING STATISTICAL LEARNING

摘要

A method for predicting natural gas demand through statistical learning is disclosed. According to an embodiment of the present invention, a method for predicting natural gas demand through statistical learning includes an exploratory data analysis step of analyzing key characteristics by collecting power plant and gas demand state information measured at a plurality of points of a gas pipe network; A state information classification step of classifying the gas demand state information collected in the exploratory data analysis step into similar groups; A classification model establishment step of establishing a model for calculating the probability of occurrence of the state information based on the state information classified in the state information classification step; And a gas demand prediction step of predicting gas demand for each power plant and time period required for system analysis by using the classification model established in the classification model establishment step. According to the present invention, it is possible to improve the efficiency, stability and national benefits of the domestic natural gas industry by supporting rational decision-making of related organizations in connection with the joint use of piping facilities through reliable prediction of power plants and gas demand by time. There is an effect that can be.