Glass production raw material proportioning prediction method

The invention discloses a glass production raw material proportioning prediction method, and relates to the technical field of glass raw material proportioning, and the method comprises the steps: collecting glass production process data of each time in the past, carrying out the quantification, normalization and vectorization of the production process data, taking the data as the input of a machine learning model, and respectively taking the gas rate and the sodium carbonate volatilization rate as prediction targets; training a machine learning model; obtaining a machine learning model for predicting the gas rate and the sodium carbonate volatilization rate; the mirabilite content, the pulverized coal content and the fluorite content are calculated by sampling mirabilite, pulverized coal and fluorite from the raw materials; in actual glass production, the gas rate and the sodium carbonate volatilization rate are calculated by setting production process data in advance, and the actual demand materials of vari