Prediction of AC Swing Head Dynamic Performance Based on TCN and SegRNN

The AC swing head, a critical component of five-axis CNC machine tools, directly affects machining accuracy and stability through its dynamic performance. This study proposes a predictive model integrating Temporal Convolutional Network (TCN) and Segmented Recurrent Neural Network (SegRNN) for high-precision forecasting of the AC swing head’s dynamic performance. The model employs TCN’s causal and dilated convolutions to extract key features from vibration velocity signals, emphasizing their impact on performance. SegRNN’s segmented processing captures short-term contextual relationships for accurate predictions. Data collected from a dedicated AC swing head experimental platform were tested, and results show the proposed method outperforms traditional models (CNN, RNN, CNN-LSTM, CNN-GRU) in Mean Squared Error (MSE) and R-squared (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) metrics, demonstrating enhanced predictive accuracy and robustness, suitable for AC swing head dynamic performance forecasting.