Machine self-learning intelligent plate changing method for edge roller

The invention discloses a machine self-learning intelligent plate changing method for an edge roller. The method comprises the following steps: inputting multiple groups of samples into a machine learning neural network; the machine learning neural network performs forward calculation according to the input sample data so as to perform model training; adjusting the weight from the input layer to the hidden layer and the weight from the hidden layer to the output layer; updating the number of times of model training, judging whether the number of times of target training is reached or not, if not, returning to the above steps to continue training, and if yes, ending training, storing trained data and obtaining a trained model; in actual production, the width and thickness of glass required by production are input into the trained model to obtain the parking space, pressing and swing angle of the glass edge roller needing to be adjusted, and the glass edge roller is adjusted to obtain glass required by production; the method has the advantages that the board changing efficiency is improved, the model is continuously optimized according to historical data, the board changing result does not depend on previously stored data, and the board changing data error is small.