国外文献 > 1.轴承 > 5. 监测与诊断 > 寿命与维护
关键词搜索：
| 标 题： |
A meta network pruning framework for remaining useful life prediction of rocket engine bearings with temporal distribution discrepancy
用于预测火箭发动机轴承剩余使用寿命的元网络剪枝框架，考虑时间分布偏差 |
| 作 者： |
Pan, Tongyang
Zhang, Sui
Li, Fudong
Chen, Jinglong
Li, Aimin |
| 关键词： |
Remaining useful life [剩余使用寿命]
Rocket engine [ 火箭发动机]
Deep learning [ 深度学习]
Attention mechanism [ 注意力机制] |
| 出版年： |
2023 |
| 期刊名： |
Mechanical Systems & Signal Processing
机械系统与信号处理 |
| 年卷期： |
2023, vol.195 |
| 摘 要： |
Accurate remaining useful life prediction (RUL) is important for the reliability and safety of liquid rocket engines. In this paper, a meta network pruning framework with attention augmented convolutions is proposed for RUL prediction. To address the problem of distribution discrepancy in engineering data under transient working conditions, a data-driven distribution matching strategy is designed. Besides, in view of the prediction accuracy and computation complexity of the model, an iterative meta network pruning algorithm, which automatically calculates the meta-gradients of each convolutional kernel according to the chain rule, is developed to identify, and then delete the unimportant connections in the framework. The method is verified on a highprecision cryogenic rocket engine bearing experiment platform under liquid nitrogen and received better performance than benchmark algorithms.
准确的剩余使用寿命（RUL）预测对液体火箭发动机的可靠性和安全性至关重要。本文提出了一种基于注意力增强卷积的元网络剪枝框架，用于RUL预测。为解决瞬态工作条件下工程数据分布不一致的问题，设计了数据驱动的分布匹配策略。此外，考虑到模型的预测精度和计算复杂度，开发了一种迭代式元网络剪枝算法，该算法根据链式法则自动计算每个卷积核的元梯度，从而识别并删除框架中不重要的连接。该方法在液氮环境下高精度低温火箭发动机轴承实验平台上得到验证，其性能优于基准算法。 |
| 中文内容为网络AI翻译，专业术语未必准确，仅供参考。 |
京公网安备 11010202009788