Liquefied Petroleum Gas Concentration Prediction Based on Regression Model

Liquefied Petroleum Gas (LPG) is an important energy source widely used in industry, agriculture and households. However, high concentrations of LPG in the air can easily cause explosions, and its widespread use has extremely high safety risks for our lives. Among the many air quality prediction methods, LPG is only one of the attribute indicators for comprehensive air quality prediction. Therefore, the prediction accuracy of these methods for LPG is not high. In this paper, we propose to use regression models to predict LPG concentrations, taking LPG as the study object and other factors in the air as reference influencing factors, and using multiple regression models and deep learning models to compare the prediction errors. Extensive experimental results show that the prediction of LPG concentration using regression models can achieve good prediction performance. Among them, a comprehensive comparison of linear regression performs the best, which verifies the validity of our idea.