【作者】 杨红瑜； 彭开香； 焦瑞华；

【Author】 YANG Hong-yu;PENG Kai-xiang;JIAO Rui-hua;Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering, University of Science and Technology Beijing;Institute of Artificial Intelligence, School of Automation and Electrical Engineering, University of Science and Technology Beijing;

【机构】 北京科技大学自动化学院工业过程知识自动化教育部重点实验室； 北京科技大学人工智能研究院；

【摘要】 基于粒子滤波(PF)的剩余寿命预测方法已在实践中被广泛使用,但对于航空航天等复杂系统利用PF建立退化模型时,其健康指标(HI)不能由监测数据直接得到,为提高预测精度,确保可靠性与安全性,本文提出一种基于数据驱动与物理模型相结合的寿命预测方法.首先,利用DBN网络构建反映健康状态的健康指标;然后,利用基于粒子滤波的各种算法进行寿命预测;最后,为评估预测方法的性能,本文开发了一款基于MATLAB的剩余寿命预测及其性能评估工具箱.该工具箱具有良好的图形用户界面,优良的人机交互性能,方便使用者进行操作.工具箱内集成了12种主要的算法,通过使用C-MAPSS航空发动机数据集进行验证,验证了该工具箱的有效性、多功能性和实用性.更多还原

【Abstract】 The remaining useful life prediction method based on particle filter(PF) has been widely used in practice. However,when using PF to establish a degradation model for aerospace and other complex systems, the health indicator(HI) cannot be obtained directly from the monitoring data. To improve prediction accuracy and ensure reliability, this paper presents a remaining useful life prediction method based on the combination of data-driven and physical models. First, the DBN network is used to construct health indicator that reflect the state of health; then, various algorithms based on particle filtering are used to predict remaining useful life; Finally, In order to evaluate the performance of the prediction methods, this article developed a MATLABbased remaining useful life prediction and its performance evaluation toolbox. The toolbox has a good graphical user interface and excellent human-computer interaction performance. So it is convenient for users to operate and twelve main algorithms are integrated in the toolbox. The C-MAPSS aircraft engine data set is used to verify the effectiveness, versatility and practicality of the toolbox.更多还原