基于改进卷积神经网络的航空发动机剩余寿命预测  

A Remaining Useful Life Prediction for Aero-Engine Based on Improved Convolution Neural Networks

作　　者：马忠 郭建胜[1] 顾涛勇 毛声 MA Zhong;GUO Jiansheng;GU Taoyong;MAO Sheng(Equipment Management and UAV Engineering College,Air Force Engineering University,Xi’an 710051,China)

机构地区：[1]空军工程大学装备管理与无人机工程学院,西安710051

出　　处：《空军工程大学学报（自然科学版）》2020年第6期19-25,共7页Journal of Air Force Engineering University(Natural Science Edition)

摘　　要：航空发动机结构复杂,状态变量多且相互之间存在着严重非线性特征,传统的基于物理失效模型的方法难以精确地预测发动机的剩余寿命(RUL)。针对此问题,采用改进的卷积神经网络(CNN)方法对发动机剩余寿命进行预测。预测过程通过建立退化模型,给每个训练样本添加RUL标签;为了更好地提取发动机使用过程中状态变量与剩余寿命之间的相关关系,使用不同的一维卷积核提取序列趋势信息特征;将特征输入构建的卷积神经网络得到剩余寿命的预测值。为了验证方法的有效性,在NASA提供的涡轮风扇发动机仿真数据集(C-MAPSS)上进行了测试,并与深度信念网络等方法对比,结果表明改进的卷积神经网络拥有更高的精度。Aimed at the problems that aero-engine is complex in structure,severe nonlinearity of various degenerate state is variable,and traditional physical failure model-based method is difficult to predict the remaining useful life of the engine(Remaining Useful Life,RUL)accurately,the problems above-mentioned can be done by adopting an improved convolution neural networks(CNN).A linear degradation model is employed to label each sample.The convolution is set to several different one-dimensional convolutions to extract data features and the correlation between the RUL better.In order to validate the effectiveness of the method,a test is made on the commercial modular aero-propulsion system simulation(C-MAPSS)aircraft engine datasets provided by NASA.The results show that the convolutional neural network has higher precision compared with the common neural network.

关 键 词：航空发动机 剩余寿命 卷积神经网络 线性退化 时间窗 

分 类 号：V23[航空宇航科学与技术—航空宇航推进理论与工程] TP183[自动化与计算机技术—控制理论与控制工程]