Grading ring parameter optimization method based on convolutional neural network

The invention discloses a grading ring parameter optimization method based on a convolutional neural network. The grading ring parameter optimization method comprises the steps of obtaining a parameter feature data set of a grading ring; the parameter feature data set is used to train a pre-constructed convolutional neural network model to obtain an electric field intensity prediction model, the convolutional neural network comprises a network feature extraction layer, a network feature sharing layer and a network feature fusion layer which are connected in sequence, and the electric field intensity prediction model is obtained. The network characteristic extraction layer comprises a hidden layer for parameter amplification; traversing in a parameter search space by adopting the electric field intensity prediction model to obtain different parameter combinations of the grading ring; evaluating the performance of each group of parameter combination of the grading ring, and determining an optimal grading ring p