Application of model based on artificial immune RBF neural network to predict silicon content in hot metal

This paper studied a Radial Basis Function(RBF) network learning algorithm based on immune recognition principle. In the algorithm, the recognized data is regarded as antigens and the compression mapping of antigens as antibodies, i, e, the hidden layer centers. In order to improve convergence speed and precision of the RBF network, we adopt the least square algorithm to determine the weights of the output layer. Applying the model to blast furnace of a large iron and steel Group Co., application result shows that the model possesses far superior forecast precision and requires less constructing time.