石慧,王婉娜,张岩,刘佳媛

• 1:太原科技大学电子与信息工程学院

摘要(Abstract)：

设备运行数据蕴含了大量的状态退化信息,利用状态退化信息对设备进行剩余寿命预测,可以为预测性维修提供重要的依据。针对监测过程中接收到表征设备退化的大量数据,提出一种改进的基于记忆机理循环神经网络实时剩余寿命预测模型。该模型利用随机搜索选择模型超参数,采用加入动量考虑的随机梯度下降算法优化模型参数,防止陷入局部最优,从而提高剩余寿命预测模型的预测精度。通过齿轮弯曲疲劳试验,验证了该模型剩余寿命预测的准确性。

关键词(KeyWords)： 剩余寿命;深度学习;LSTM;动量参数

Abstract:

Keywords:

基金项目(Foundation): 国家青年科学基金项目(61703297);; 校博士启动基金(20152022)

作者(Author): 石慧,王婉娜,张岩,刘佳媛

参考文献(References)：

• [1] LIU Z,ZHEN J,VONG C M,et al. A Patent Analysis of Prognostics and Health Management(PHM)Innovations for Electrical Systems[J]. IEEE Access,2018,6(99):18088-18107.
• [2] GRAHAM-ROWE D,GOLDSTON D,DOCTOROW C,et al. Big data:Science in the petabyte era[J]. Nature,2008,455(7209):8-9.
• [3] TOBON-MEJIA D A,MEDJAHER K,ZERHOUNI N,et al. A Data-Driven Failure Prognostics Method Based on Mixture of Gaussians Hidden Markov Models[J]. IEEE Transactions on Reliability,2012,61(2):491-503.
• [4]郭远晶，魏燕定，金晓航，等．基于S变换谱核密度估计的齿轮故障诊断[J]．仪器仪表学报，2017,38(6):1432-1439.
• [5]刘颖，严军．基于时间序列ARMA模型的振动故障预测[J]．化工自动化及仪表，2011,38(7):841-843.
• [6] WANG X,WU J,LIU C,et al. A Hybrid Model Based on Singular Spectrum Analysis and Support Vector Machines Regression for Failure Time Series Prediction[J]. Quality&Reliability Engineering International,2016,32(8):2717-2738.
• [7] DAVIES T M,FLYNN C R,HAZELTON M L. On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation[J]. Statistics&Probability Letters,2018,138:75-81.
• [8] LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521(7553):436.
• [9] SCHMIDHUBER J,RGEN. Deep learning in neural networks[M]. Elsevier Science Ltd,2015.
• [10] FARDIS N,MEHDI I. Application and comparison of RNN,RBFNN and MNLR approaches on prediction of flotation column performance[J]. International Journal of Mining Science and Technology,2015,25(6):983-990.
• [11] DOETSCH P,KOZIELSKI M,NEY H. Fast and Robust Training of Recurrent Neural Networks for Offline Handwriting Recognition[C]//International Conference on Frontiers in Handwriting Recognition. IEEE,2014:279-284.
• [12] HAITAO ZHAO,SHAOYUAN SUN,BO JIN. Sequential Fault Diagnosis Based on LSTM Neural Network[J]. IEEE Access,2018,6:12929-12939.
• [13]赵建鹏，周俊．基于长短时记忆网络的旋转机械状态预测研究[J]．噪声与振动控制，2017,37(4):155-159.
• [14]王鑫，吴际，刘超，等．基于LSTM循环神经网络的故障时间序列预测[J]．北京航空航天大学学报，2018,44(4):772-784.
• [15] HOCHREITER S,SCHMIDHUBER J. Long Short-Term Memory[J]. Neural Computation,1997,9(8):1735-1780.
• [16] ELSAID A E,WILD B,HIGGINS J,et al. Using LSTM recurrent neural networks to predict excess vibration events in aircraft engines[C]//2016 IEEE 12th International Conference on e-Science(e-Science). IEEE Computer Society,2016:260-269.
• [17] JIANJING Z,PENG W,RUQIANG Y,et al. Long short-term memory for machine remaining life prediction[J]. Journal of Manufacturing Systems,2018,48:78-86.
• [18] DUCHI J,HAZAN E,SINGER Y. Adaptive Subgradient Methods for Online Learning and Stochastic Optimization[J]. Journal of Machine Learning Research,2011,12(7):257-269.
• [19] HINTON G E,SRIVASTAVA N,KRIZHEVSKY A,et al. Improving neural networks by preventing co-adaptation of feature detectors[J]. Computer Science,2012,3(4):212-223.
• [20] ARCOS GARCIA,ALVARO,ALVAREZ GARCIA,et al. Deep neural network for traffic sign recognition systems:An analysis of spatial transformers and stochastic optimisation methods[J]. Neural Networks,2018,99(3):158-169.
• [21] BERGSTRA J,BENGIO Y. Random search for hyper-parameter optimization[J]. Journal of Machine Learning Research,2012,13(1):281-305.
• [22]梁泽明，姜洪权，周秉直，等．多参数相似性信息融合的剩余寿命预测[J]．计算机集成制造系统，2018,24(4):813-819.