Mold processing quality prediction based on TCN-DANN adaptive transfer

CNC machining is a key process for achieving complex cavities, high surface quality, and precise fits. However, in new scenarios, its machining data often faces challenges such as scarce labeled samples and noise contamination in long-term sequential data, limiting the performance of quality prediction models. Therefore, a mold processing quality prediction method based on TCN-DANN adaptive transfer learning has been proposed. Firstly, features are constructed and quality index is defined based on CNC machining data. Secondly, a temporal convolution network (TCN) is utilized to extract deep-level features from noisy data, and an improved domain adversarial neural network (DANN) is integrated to achieve quality prediction through cross-domain adversarial learning, thereby enhancing the model’s generalization capability under varying working conditions. Finally, a transfer learning task is designed to conduct ablation experiments. The results show that the proposed method performs excellently in cross-condition quality prediction, validating the effectiveness of its prediction performance and generalization ability.