Remaining Useful Life Prediction of End Mills Using DCNN-McBiLSTM-LRSA with Multi-Source Sensory Signals

In precision mold manufacturing, the machining of HRC52 hardened steel causes se-vere tool wear and high noise in multi source sensor signals, making accurate remain-ing useful life (RUL) prediction challenging. To address this, we propose a hybrid mod-el that integrates one dimensional deep convolution (DCNN), low resolution self attention (LRSA) with 1D 2D spatiotemporal reconstruction, and a multi channel bidirectional long short term memory network (McBiLSTM). A Gaussian smoothing filter is first applied to denoise the 50 kHz signals, followed by physical period sliding windows for feature extraction. A multi strategy fusion pooling layer (mean, max, and last quarter features) further improves prediction accuracy. Using the PHM 2010 milling cutter dataset under leave one out cross validation, the proposed model achieves a mean absolute percentage error (MAPE) of 1.45% and a root mean square error (RMSE) of 2.76 mm, reducing prediction error by up to 75.6% compared to Transformer, LSTM, and GRU baselines. These results demonstrate that the model ef-fectively extracts degradation features even during the accelerated wear stage, offer-ing a reliable solution for tool health monitoring and predictive maintenance under complex cutting conditions.