Bearing Health Condition Prediction Using Deep Belief Network
Bearing Health Condition Prediction Using Deep Belief Network
复制标题
DOI：
10.36001/phmconf.2017.v9i1.2484
复制
发表时间：
2017-10
期刊：
影响因子：
--
通讯作者：
Guangquan Zhao;Xiaoyong Liu;Bin Zhang;Guohui Zhang;Guangxing Niu;Cong Hu
中科院分区：
文献类型：
--
作者：
Guangquan Zhao;Xiaoyong Liu;Bin Zhang;Guohui Zhang;Guangxing Niu;Cong Hu
轴承在维持旋转机械的安全性和可靠性方面起着至关重要的作用。轴承健康状态预测为状态维修提供决策信息，目的是防止意外故障的发生，最大限度地降低总体维修成本。提出了一种基于深度信念网络(DBN)的数据驱动轴承健康状态预测方法。在该预测方法中，使用DBN作为预测器，它包括堆叠的RBM和回归输出。我们的主要贡献包括开发基于深度学习的数据驱动预测解决方案，该解决方案不依赖显式模型方程和预测专业知识，并提供对五个典型运行到故障轴承的全面预测结果。使用IEEE PHM 2012挑战数据集验证了该方法的有效性，并与现有的两种方法进行了比较。结果表明，该方法在轴承的短期健康状态预测和剩余使用寿命预测方面具有良好的性能。
Bearings play a critical role in maintaining safety and reliability of rotating machinery. Bearings health condition prediction aims to prevent unexpected failures and minimize overall maintenance costs since it provides decision making information for condition-based maintenance. This paper proposes a Deep Belief Network (DBN)-based data-driven health condition prediction method for bearings. In this prediction method, a DBN is used as the predictor, which includes stacked RBMs and regression output. Our main contributions include development of a deep leaning-based data-driven prognosis solution that does not rely on explicit model equations and prognostic expertise, and providing comprehensive prediction results on five representative run-to-failure bearings. The IEEE PHM 2012 challenge dataset is used to demonstrate the effectiveness of the proposed method, and the results are compared with two existing methods. The results show that the proposed method has promising performance in terms of short-term health condition prediction and remaining useful life prediction for bearings.