Neural network rolling force forecasting method based on K-means clustering algorithm

The invention discloses a prediction method for predicting rolling force through a neural network based on a K-means clustering algorithm. The method belongs to the technical field of computers, and specifically comprises the following steps: determining input and output layers of an RBF neural network; estimating the number of input and output layers and hidden nodes of the nonlinear multilayer forward RBF neural network; a hidden layer space is formed; determining a proper data center, and determining an expansion constant of a hidden node according to the distance between the centers; training the artificial neural network, learning and correcting errors, and completing the construction of the artificial neural network; and the rolling force is preset by adopting the artificial neural network for production. Compared with a traditional nonlinear multilayer forward neural network, the method has the advantages that the operation speed is high, the model is easy to maintain, meanwhile, the defects that the n