镍基高温合金铣削刀具磨损预测

Tool Wear Prediction for Nickel-Based Superalloy Milling

搭建镍基高温合金铣削实验测试平台,分析刀具磨损变化规律,提出了一种基于堆叠稀疏自动编码器和多传感器特征融合的新型深度学习方法,用于铣削刀具磨损预测。在时域、频域和时频域中提取信号特征,并通过相关性分析确定最优的多传感器特征,输入堆叠稀疏自动编码器进行深度特征学习。利用双向长短时记忆网络建立刀具磨损预测模型,应用不同的铣削磨损实验数据集来验证训练模型的预测性能。预测结果表明,所提模型均方根误差与传统模型相比至少减小了9.6%,证明了多传感器特征融合和深度学习方法的结合可以提高预测性能。

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.