Abstract：A new deep learning method based on stacked sparse autoencoders and multi-sensor feature fusion is proposed for milling tool wear prediction by building a nickel-based high temperature alloy milling experimental test platform and analysing tool wear variation patterns. Signal features are extracted in the time domain,frequency domain and time-frequency domain,and the optimal multi-sensor features are determined through correlation analysis,which is input to SSAE for deep feature learning. A tool wear prediction model is established using a bidirectional long-short term memory,and different experimental data sets of milling wear are applied to verify the prediction performance of the trained model. The prediction results show that the root-mean-square error is reduced by at least 9.6% compared to each of the known models, proving that the combination of multi-sensor feature fusion and deep learning methods can improve the prediction performance.
He Y H. Turning of Nickel Base Alloy［J］. Metal Working(Metal Cutting), 2010(9): 36-37.
Zhang K F, Yuan H Q, Nie P. Tool wear state monitoring based on generalized fractal dimension ［J］. Vibration and Shock, 2014, 33(1): 162-164.
Luo H, Zhang D H, Luo M. Research progress of cutting tool wear and remaining life prediction of difficult-to-machine materials in aviation［J］. China Mechanical Engineering, 2021, 32(22): 2647-2666.
Wang Q, Li Y G, Hao X Z, et al. Dynamic prediction method of CNC machining tool life based on online learning［J］. Aeronautical Manufacturing Technology, 2019, 62(21): 22-30.
Wu F H, Zhong H, Zhang Q, et al. Tool Wear Condition Monitoring Based on Convolution-gated Recurrent Neural Network［J］. Acta Metrologica Sinica, 2021, 42(8): 1034-1040.
Huang Z P, Huang X Y, Li L, et al. Research on data-driven milling tool life prediction［J］. Manufacturing Technology & Machine Tool, 2020 (1): 153-161.
Li H J, Kang W M, Su J D. Research on machining of nickel-based superalloys［J］. CFHI Technology, 2017 (2): 68-71.
Huang K. Study on Machinability of Nickel Base Superalloy ［J］. Mechanical Engineer, 2010 (6): 160-161.
Wu X F, Chen J F, Yin X F, et al. GA-BP neural network tool life prediction based on laser heating-assisted milling［J］. Tool Engineering, 2018, 52(7): 57-61.
An H, Wang G F, Wang Z, et al. Tool condition monitoring and remaining life prediction method based on deep learning theory［J］. Journal of Electronic Measurement and Instrumentation, 2019, 33(9): 64-70.
［13］
Tao Z, An Q, Liu G, et al. A Novel Method for Tool Condition Monitoring Based on Long Short-term Memory and Hidden Markov Model Hybrid Framework in High-speed Milling Ti-6Al-4V［J］. International Journal of Advanced Manufacturing Technology, 2019, 105(7/8): 3165-3182.
He Z J, Zhou Z X, Huang X M. Tool Wear State Monitoring Based on Variational Mode Decomposition and Correlation Dimension and Relevance Vector Machine［J］. Acta Metrologica Sinica, 2018, 39(2): 182-186.
Guo B S, Han T J, Zhang Y, et al. Tool Wear Monitoring Method Based on One-Dimensional Residual Network with Threshold Module［J］. Acta Metrologica Sinica, 2022, 43(4): 501-506.
Luo J, Tong J Y, Zheng J D, et al. Fault diagnosis method for rolling bearings based on EEMD and stacked sparse self-coding ［J］. Noise and Vibration Control, 2020 (2): 115-120.
［12］
Kong D, Chen Y, Li N, et al. Tool wear estimation in end milling of Titanium alloy using NPE and a novel WOA-SVM model［J］. IEEE Transactions on Instrumentation and Measurement, 2020, 69(7): 5219-5232.
［14］
Wang P, Liu Z, Gao R X, et al. Heterogeneous Data-driven Hybrid Machine Learning for Tool Condition Prognosis［J］. CIRP Annals, 2019, 68(1): 455-458.
2023, Vol. 44 
