Parameters Predictions of Paddy Grain Drying Based on Machine Learning

In order to improve the drying efficiency and guarantee the grain quality after drying, this paper proposed to use artificial intelligence modelling to predict the grain quality indexes after drying. With air temperature, air relative humidity, initial moisture content, air velocity and tempering rate as control parameters, three models were established, and the prediction performance of these three models was tested. The three models were regression model, Back-propagation Neural Network (BPNN) and Deep Neural Network (DNN). The results showed that the machine learning techniques, in particular DNN had the best performance, especially for predicting germination rate ratio. The contribution of this paper is to demonstrate the ability of machine learning techniques in fitting grain drying characteristics with different control parameters. This paper proved the applicability of machine learning technology in grain drying field, and established the corresponding models.