针对传统寿命预测方法计算复杂、耗时且不具普适性等问题,提出一种基于优化长短期记忆网络（LSTM）的风力机叶片剩余使用寿命（RUL）预测模型。首先,将多维传感器监测数据可视化,以观察数据特征并进行初次特征筛选。然后,对筛选后的数据进行归一化处理,并使用主成分分析法（PCA）进行数据融合,以去除冗余信息和降低特征维度。其次,使用自适应矩估计（AME）算法为不同网络参数提供独立的自适应性学习率;使用平滑平均绝对误差（SMAE）损失函数来综合两种传统回归损失函数的特点。最后,经过多次试验选定合适的LSTM层数及神经元数,并以复杂系统的多尺度时序监测数据为算例对模型进行试验验证。试验结果表明,在一种故障模式下,优化LSTM预测模型相较于其他传统机器学习模型在评价指标及预测误差分布情况上占优,表明该文所提模型具有更高的准确性及稳定性。
Abstract
Aiming at the problems of complex calculation, time consuming and inapplicability of traditional life prediction methods, a wind turbine blade remaining useful life （RUL） prediction model based on optimized Long Short-Term Memory （LSTM） is proposed. In this study, the multidimensional sensor monitoring data were visualized to observe the data features and perform initial feature screening. Then, the filtered data were normalized and the data were fused using principal component analysis （PCA） to remove redundant information and reduce feature dimensionality. Furthermore, the adaptive moment estimation （AME） algorithm was employed to provide independent adaptive learning rates for different network parameters, and the smoothed mean absolute error （SMAE） loss function was utilized to synthesize the characteristics of two traditional regression loss functions. After several experiments, the optimal number of LSTM layers and neurons was selected. The model was experimentally validated using multi-scale time-series monitoring data of complex systems as an arithmetic example. The experimental results demonstrate that the optimized LSTM prediction model outperforms other traditional machine learning models in terms of evaluation index and prediction error distribution under one fault mode. This indicates that the proposed model offers higher accuracy and stability.
关键词
风力机叶片 /
主成分分析 /
长短期记忆 /
寿命预测 /
预测模型
Key words
wind turbine blades /
principal component analysis /
Long short-term memory /
life prediction /
prediction model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] EL-THALJI I, JANTUNEN E.On the development of condition based maintenance strategy for offshore wind farm: requirement elicitation process[J]. Energy procedia, 2012, 24: 328-339.
[2] CHOU J S, TU W T.Failure analysis and risk management of a collapsed large wind turbine tower[J]. Engineering failure analysis, 2011, 18(1): 295-313.
[3] 朱朔, 白瑞林, 吉峰. 改进CHSMM的滚动轴承剩余寿命预测方法[J]. 机械传动, 2018, 42(10): 46-52, 95.
ZHU S, BAI R L, JI F.Rolling bearing remaining useful life prognosis method based on improved CHSMM[J]. Journal of mechanical transmission, 2018, 42(10): 46-52, 95.
[4] 安宗文, 杨晓玺, 寇海霞, 等. 1.5 MW风电叶片多轴疲劳寿命分析[J]. 太阳能学报, 2020, 41(5): 129-135.
AN Z W, YANG X X, KOU H X, et al.Multi-axial fatigue life analysis of 1.5 MW wind turbine blades[J]. Acta energiae solaris sinica, 2020, 41(5): 129-135.
[5] 姜媛媛, 曾文文, 沈静静, 等. 基于凸优化-寿命参数退化机理模型的锂离子电池剩余使用寿命预测[J]. 电力系统及其自动化学报, 2019, 31(3): 23-28.
JIANG Y Y, ZENG W W, SHEN J J, et al.Prediction of remaining useful life of lithium-ion battery based on convex optimization-life parameter degradation mechanism model[J]. Proceedings of the CSU-EPSA, 2019, 31(3): 23-28.
[6] AGHAJANI S, HEMATI M, TORABNIA S.Life prediction of wind turbine blades using multi-scale damage model[J]. Journal of reinforced plastics and composites, 2021, 40(17/18): 644-653.
[7] 段桂英, 姜洪开. 基于数据融合驱动和DLSTM网络的轴承RUL预测[J]. 计算机应用与软件, 2021, 38(12): 22-29.
DUAN G Y, JIANG H K.Bearing RUL prediction based on data fusion drive and DLSTM network[J]. Computer applications and software, 2021, 38(12): 22-29.
[8] 马奇友, 刘可薇, 杜坚, 等. 基于深度长短期记忆网络的发动机叶片剩余寿命预测[J]. 推进技术, 2021, 42(8): 1888-1897.
MA Q Y, LIU K W, DU J, et al.Prediction of residual life of engine blades based on deep short term memory network[J]. Journal of propulsion technology, 2021, 42(8): 1888-1897.
[9] XIAO L, ZHANG L Y, NIU F, et al.RETRACTED: remaining useful life prediction of wind turbine generator based on 1D-CNN and Bi-LSTM[J]. International journal of fatigue, 2022, 163: 107051.
[10] HOCHREITER S, SCHMIDHUBER J.Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[11] 王颖洁, 朱久祺, 汪祖民, 等. 自然语言处理在文本情感分析领域应用综述[J]. 计算机应用, 2022, 42(4): 1011-1020.
WANG Y J, ZHU J Q, WANG Z M, et al.Review of applications of natural language processing in text sentiment analysis[J]. Journal of computer applications, 2022, 42(4): 1011-1020.
[12] 贾澎涛, 孙炜. 基于深度学习的文本分类综述[J]. 计算机与现代化, 2021(7): 29-37.
JIA P T, SUN W.A survey of text classification based on deep learning[J]. Computer and modernization, 2021(7): 29-37.
[13] 王雨嫣, 廖柏林, 彭晨, 等. 递归神经网络研究综述[J]. 吉首大学学报(自然科学版), 2021, 42(1): 41-48.
WANG Y Y, LIAO B L, PENG C, et al.Research review of recurrent neural networks[J]. Journal of Jishou University (natural sciences edition), 2021, 42(1): 41-48.
[14] KINGMA D P, BA J. Adam: a method for stochastic optimization[EB/OL].2014: arXiv: 1412.6980. http://arxiv.org/abs/1412.6980.pdf
[15] GIRSHICK R.Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision (ICCV). Santiago, Chile, 2015: 1440-1448.
[16] ZHAO Z B, WU J Y, LI T F, et al.Challenges and opportunities of AI-enabled monitoring, diagnosis \& prognosis: a review[J]. Chinese journal of mechanical engineering, 2021, 34(1): 56.
[17] 李彦夫, 韩特. 基于深度学习的工业装备PHM研究综述[J]. 振动测试与诊断, 2022, 42(5): 835-847, 1029.
LI Y F, HAN T.Deep learning based industrial equipment prognostics and health management: a review[J]. Journal of vibration, measurement & diagnosis, 2022, 42(5): 835-847, 1029.
[18] 栗然, 马涛, 张潇, 等. 基于卷积长短期记忆神经网络的短期风功率预测[J]. 太阳能学报, 2021, 42(6): 304-311.
LI R, MA T, ZHANG X, et al.Short-term wind power prediction based on convolutional long-short-term memory neural networks[J]. Acta energiae solaris sinica, 2021, 42(6): 304-311.
[19] 王依宁, 解大, 王西田, 等. 基于PCA-LSTM模型的风电机网相互作用预测[J]. 中国电机工程学报, 2019, 39(14): 4070-4081.
WANG Y N, XIE D, WANG X T, et al.Prediction of interaction between grid and wind farms based on PCA-LSTM model[J]. Proceedings of the CSEE, 2019, 39(14): 4070-4081.
[20] YUAN M, WU Y T, LIN L.Fault diagnosis and remaining useful life estimation of aero engine using LSTM neural network[C]//2016 IEEE International Conference on Aircraft Utility Systems (AUS), Beijing, China, 2016: 135-140.
[21] ZHENG S, RISTOVSKI K, FARAHAT A, et al.Long Short-Term Memory Network for Remaining Useful Life estimation[C]//2017 IEEE International Conference on Prognostics and Health Management (ICPHM), Dallas, TX, USA, 2017: 88-95.
[22] MATHEW V, TOBY T, SINGH V, et al.Prediction of Remaining Useful Lifetime (RUL) of turbofan engine using machine learning[C]//2017 IEEE International Conference on Circuits and Systems(ICCS). Thiruvananthapuram, India, 2017: 306-311.
基金
国家自然科学基金（52165019）; 内蒙古自治区直属高校基本科研业务费项目（JY20220305）
PDF(1971 KB)
