With the Chinese data for solid rocket engines, traditional named entity recognition cannot be used to learn both character features and contextual sequence-related information from the input text, and there is a lack of research on the advantages of dual-channel networks. To address this problem, this paper proposes a BERT-based dual-channel named entity recognition model for solid rocket engines. This model uses a BERT pre-trained language model to encode individual characters, obtaining a vector representation corresponding to each character. The dual-channel network consists of a CNN and BiLSTM, using the convolutional layer for feature extraction and the BiLSTM layer to extract sequential and sequence-related information from the text. The experimental results showed that the model proposed in this paper achieved good results in the named entity recognition task using the solid rocket engine dataset. The accuracy, recall and F1-score were 85.40%, 87.70% and 86.53%, respectively, which were all higher than the results of the comparison models.