期刊
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
卷 69, 期 9, 页码 9451-9461出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIE.2021.3112987
关键词
Logic gates; Servomotors; Predictive models; Adaptive systems; Adaptation models; Feature extraction; Degradation; Attention hierarchy mechanism; health indicator (HI); long-term prediction; remaining useful life (RUL) prediction; servo motor
资金
- National Natural Science Foundation of China [52175075, 62033001]
- Graduate Scientific research and Innovation Foundation of Chongqing, China [CYB19007]
This article introduces a novel variant of recurrent neural networks, called gated adaptive hierarchical attention unit (GAHAU), for efficient prediction of the remaining useful life (RUL) of servo motors. The GAHAU not only pays attention to the outputs of reset and update gates, but also adaptively divides the hierarchy of sequence information in the attention gate. Experimental results demonstrate that the presented GAHAU has remarkable prediction performance, especially for long-term prediction.
The remaining useful life (RUL) prediction of servo motors plays a greatly indispensable role in preventing abnormal and inefficient operation of servo drive system and even major safety accident. In this article, a novel variant of recurrent neural networks, called gated adaptive hierarchical attention unit (GAHAU), is constructed to efficiently predict the RULs of servo motors, which can not only pay attention to the outputs of reset and update gates but also adaptively divide the hierarchy of the sequence information in attention gate. The updating formulas of GAHAU are derived. With the raw vibration signals of servo motors, a quadratic function-based deep convolutional autoencoder is used for automatically generating the health indicator vector. Based on the constructed HI vector, the RULs of servo motors are accurately predicted by GAHAU. The experiments on servo motors demonstrate that the presented GAHAU has remarkable prediction performance especially for the long-term prediction. Compared to the existing typical RUL prediction methods, the proposed methodology has stronger predictive ability according to the comparative results.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据