Research on Fermented Grains Steam Running Condition Prediction Model of Make Chinese Liquor Robot Based on GRU

Abstract

This specification is set for the theses to be published in Computer Applications and Software, including fonts, margins, page size and print area. Distillation is the key step in Chinese liquor brewing, accurate prediction of fermented grains Steam Running Condition plays an important role in the process of accurate gas exploration, when there are several areas on the surface of fermented grains running off, the robot on the steamer cannot complete the spreading operation in time, which is easy to cause wine loss and directly affect the quality of wine. To solve this problem, Fermented Grains Steam Running Condition Prediction Model of Make Wine Robot Based on GRU recurrent neural network was proposed. It used historical Steam Running Condition information of fermented grains and Hough transform to extract the Steam Running Condition data of fermented grains and find out the important factors that affect the Steam Running Condition of fermented grains. The Bayesian optimization algorithm is used to search for the optimal parameters, we built GRU Steam Running Condition Prediction Model to achieve accurate prediction of Steam Running Condition. We used the relevant data of the steaming process in Jinpai distillery. The results show that the model can better predict the variation trend of Fermented Grains Steam Running Condition.