Incremental learning BiLSTM based on dynamic proportional adjustment mechanism and experience replay for quantitative detection of blade crack propagation

Abstract

In the traditional quantitative detection model for blade cracks in centrifugal fan, it is assumed that the data distribution is fixed or stable. However, the new data brought by the crack propagation would break the stable distribution, thereby disturbing the old data, and resulting in a decrease in the detection performance of the model. To overcome catastrophic forgetting and reduce the extra computational cost of retaining intact old data, a quantitative detection method based on incremental learning bidirectional long short-term memory (BiLSTM) with dynamic proportional adjustment mechanism and experience replay for blade crack propagation is proposed. First, a basic BiLSTM model is constructed by inputting the data of cracks with a length of 0–5 mm. Second, the fully connected layer features in the model are selected for t-distributed stochastic neighbor embedding (t-SNE) dimensional reduction, and the Kullback–Leibler divergence is used as an indicator of feature distribution evaluating the representative old data. Third, a dynamic proportional adjustment mechanism for the old data retention proportion is constructed according to the feature distribution index and the model detection accuracy. Finally, the data of the crack with a length of 6–10 mm are gradually input to proceed with the incremental learning of the model. Verified by the measured data of the centrifugal fan, the model can adjust the retained number of old crack length data dynamically, and import new crack length data for incremental learning, making it characterized by high detection accuracy, stability, and plasticity for the quantitative detection of crack length propagation in blades.

Keywords

Incremental learning BiLSTM experience replay dynamic proportional adjustment mechanism blade crack detection

Get full access to this article

View all access options for this article.

References

Song

. Crack damage monitoring for compressor blades based on acoustic emission. Struct Health Monit 2022; 21(6): 2641–2656.

Gai

Shen

, et al. An integrated method based on hybrid grey wolf optimizer improved variational mode decomposition and deep neural network for fault diagnosis of rolling bearing. Measurement 2020; 162: 107901.

Zhang

Jiang

Feng

. Research on variational mode decomposition in rolling bearings fault diagnosis of the multistage centrifugal pump. Mech Syst Signal Process 2017; 93: 460–493.

Zhao

. Broad convolutional neural network based industrial process fault diagnosis with incremental learning capability. IEEE Trans Ind Electron 2020; 67: 5081–5091.

Tang

, et al. Fault diagnosis method based on incremental enhanced supervised locally linear embedding and adaptive nearest neighbor classifier. Measurement 2014; 48: 136–148.

Zou

Xia

. Fault Diagnosis of Tennessee-Eastman process using orthogonal incremental extreme learning machine based on driving amount. IEEE Trans Cybern 2018; 48: 3403–3410.

Rebuffi

Kolesnikov

Sperl

, et al. iCaRL: incremental classifier and representation learning. IEEE Computer Society, Honolulu, HI, USA, 21–26 July 2017. New York: IEEE.

David

Arun

Jonathan

, et al. Experience replay for continual learning. NeurIPS 2019; 2019: 1811.11682.

Shieh

Haq

, et al. Continual learning strategy in one-stage object detection framework based on experience replay for autonomous driving vehicle. Sensors 2020; 20: 6777.

10.

Wang

Subakan

Tzinis

, et al. Continual learning of new sound classes using generative replay. IEEE WASPAA 2019; 2019: 308–312.

11.

Hochreiter

Schmidhuber

. Long short-term memory. Neural Comput 1997; 9: 1735–1780.

12.

Tang

Liu

, et al. A semi-supervised transferable LSTM with feature evaluation for fault diagnosis of rotating machinery. Appl Intell 2021; 52: 1703–1717.

13.

Zhang

Niu

Zhang

, et al. Cost-effective Lebesgue sampling long short-term memory networks for lithium-ion batteries diagnosis and prognosis. IEEE Trans Ind Electron 2021; 69: 1958–1967.

14.

Zou

Zhang

Sang

, et al. Bearing fault diagnosis based on combined multi-scale weighted entropy morphological filtering and bi-LSTM. Appl Intell 2021; 51: 6647–6664.

15.

Liu

Yang

, et al. Sequence fault diagnosis for PEMFC water management subsystem using deep learning with t-SNE. IEEE Access 2019; 7: 92009–92019.

16.

Dai

Zhou

. Short-term traffic flow prediction for urban road sections based on time series analysis and LSTM_BILSTM method. IEEE Trans. Intell Transp Syst 2021; 23: 5615–5624.

17.

Bounoua

Benkara

Kouadri

, et al. Online monitoring scheme using principal component analysis through Kullback–Leibler divergence analysis technique for fault detection. Trans Inst Meas Control 2019; 42: 1225–1238.

18.

Harmouche

Delpha

Diallo

, et al. Statistical approach for nondestructive incipient crack detection and characterization using Kullback–Leibler divergence. IEEE Trans Reliab 2016; 63: 1360–1368.

19.

Zhao

Wang

, et al. Chaos-enhanced synchronized bat optimizer. Appl Math Model 2019; 77: 1201–1215.

20.

Zhou

Wang

Zhang

, et al. A novel image encryption cryptosystem based on true random numbers and chaotic systems. Multimedia Syst 2022, 28: 95–112.

21.

Zhang

Ding

. A stochastic configuration network based on chaotic sparrow search algorithm. Knowledge-Based Syst 2021; 220: 106924.

22.

Wang

Yang

, et al. A new fault diagnosis method based on adaptive spectrum mode extraction. Struct Health Monit 2021; 20: 3354–3370.

23.

Huang

Gang

Qin

. Boosting for transfer learning from multiple data sources. Pattern Recognit Lett 2012; 33: 568–579.

24.

Joss

Müller

. Machine learning for fluid property correlations: classroom examples with MATLAB. J Chem Educ 2020; 483: 229108.