Scholarlink

<p>As the executive part of metal processing, tool wear is closely related to the surface quality and machining accuracy of the workpiece. In recent years, many RUL prediction methods based on deep learning have emerged, but these methods still have some shortcomings, such as high model complexity, long running time, or because of the complexity of the processing system, using only a single sensor signal or a single model makes RUL prediction accuracy low. To overcome the above problems, this paper proposes a tool RUL prediction model based on GRU-LSTM (gated cycle unit long and short-term memory neural network). Firstly, the raw signal is sampled and processed, input into GRU for feature extraction, and then, the output of GRU is used as the input of LSTM for RUL prediction. The effectiveness of the proposed model is demonstrated by comparing it with traditional GRU and LSTM. The experimental results show that compared to traditional GRU and LSTM, the proposed model has the highest performance improvement of 53.12%.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent RUL prediction method of cutting tools based on GRU-LSTM

  • 期刊论文

  • 发表日期：

    2025年5月3日

查看全文
  • Changfu Liu,
  • Yu Quan,
  • Yang Zhou,
  • Xinli Yu,
  • Xiaoning Sun

摘要

As the executive part of metal processing, tool wear is closely related to the surface quality and machining accuracy of the workpiece. In recent years, many RUL prediction methods based on deep learning have emerged, but these methods still have some shortcomings, such as high model complexity, long running time, or because of the complexity of the processing system, using only a single sensor signal or a single model makes RUL prediction accuracy low. To overcome the above problems, this paper proposes a tool RUL prediction model based on GRU-LSTM (gated cycle unit long and short-term memory neural network). Firstly, the raw signal is sampled and processed, input into GRU for feature extraction, and then, the output of GRU is used as the input of LSTM for RUL prediction. The effectiveness of the proposed model is demonstrated by comparing it with traditional GRU and LSTM. The experimental results show that compared to traditional GRU and LSTM, the proposed model has the highest performance improvement of 53.12%.