The present invention relates to the field of automated production, and discloses a gas injection control system, which uses NARX neural network model 1 and NARX neural network model 2 to control the gas flow and predict the gas flow value respectively, because the NARX neural network introduces a delay module and The dynamic recursive network of the output feedback model, which introduces the input and output vector delay feedback into the network training, forms a new input vector, and has good nonlinear mapping ability. The input of the NARX neural network includes not only the error of the original gas flow, The input data of the control amount and the actual gas flow also includes the corresponding output data after training, and the generalization ability of the network is improved, so that it has better prediction accuracy and accuracy than the traditional static neural network in the prediction of the corresponding parameters of the gas flow. Adaptive capability, the single-chip controller improves the accuracy, robustness and system reliability of the control system.