106 / 2021-07-23 11:36:15
Multi-feature fused bidirectional long short-term memory for remaining useful life prediction
Final Paper
Ruibing Jin / The Agency for Science, Technology and Research
Zhenghua Chen / The Agency for Science, Technology and Research
Keyu Wu / The Agency for Science, Technology and Research
Min Wu / The Agency for Science, Technology and Research
Xiaoli Li / The Agency for Science, Technology and Research
如强 严 / 西安交通大学
In industry, prognostic health management (PHM) is very important to improve the system reliability and efficiency. In PHM, remaining useful life (RUL) prediction plays a necessary role in preventing machine failure and lowering operation cost. Recently, benefitted from deep learning technology development, many RUL prediction approaches based on long short-term memory (LSTM) and convolutional neural networks (CNN) are proposed and show impressive performances. However, existing deep learning based methods directly utilize raw signals. Affected by noise in the raw input, the feature representation is degraded, further degenerating the performance. To address this issue, we propose a multi-feature fused bidirectional LSTM (MF-LSTM). Our proposed MF-LSTM consists of two part: multi-feature fusion (MF) module and multi-head attentive fusion (MA) module. In MF module, feature extracted by a bidirectional LSTM is combined with traditional handcrafted features. A fusion layer is proposed in MF module, which effectively combines both features and improves the feature representation. Furthermore, an attention module is proposed according to multi-head attention mechanism, which improves the performance further. To verify our MF-LSTM performance, experiments are carried out on the C-MAPSS dataset, showing a state-of-the-art performance.
Important Date
  • Conference Date

    Oct 21



    Oct 23


  • Oct 26 2021

    Registration deadline

Sponsored By
Southeast University, China
Contact Information