Abstract
Aircraft hydraulic systems can function very competently in challenging in-flight circumstances and are used throughout the functioning of critical flight components. Health monitoring of the hydraulic system parameters is the need of the hour. The objective of this paper is to investigate and propose a noise-invariant machine learning model for health monitoring of the hydraulic system using multivariate sensor parameter time series data. Identifying the health of the hydraulic system from sampled sensor parameter values is modeled as a multivariate time series classification problem. Our contribution is twofold: 1. identification of highly significant statistical, spectral and temporal features specific to the sensor parameter value. These feature vectors are mapped to a low-dimensional feature space using a feature selector package using Benjamini Hochberg process. 2. Train and build a machine learning model highly robust and invariant to various levels of random, squeal and impulse noise. In this paper, we evaluate and examine the models performance on the noise-free original and noise-induced multivariate time series data. Experimental results show that identified and selected statistical features for each sensor parameter are more robust to low, medium and high noise levels compared to spectral features and temporal features.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Kadous, M.W., et al.: Classification of multivariate time series and structured data using constructive induction. Mach. Learn. 58, 179–216 (2005)
Liu, X., et al.: Study on knowledge-based intelligent fault diagnosis of hydraulic system. TELKOMNlKA 10, 2041–2046 (2012)
Hsieh, T.-Y., et al.: Explainable multivariate time series classification: a deep neural network which learns to attend to important variables as well as time intervals. In: WSDM’21: Proceedings of ICWSDM, pp. 607–615 (2021)
Esmael, B., et al.: Multivariate time series classification by combining trend-based and value-based approximations. In: Computational Science and Its Applications—ICCSA (2012)
Weng, X., Qin, S.: Classification of multivariate time series using supervised isomap. In: 2012 Third Global Congress on Intelligent Systems, Feb 2013, pp. 136–139. https://doi.org/10.1109/GCIS.2012.31
Xu, H., et al.: Multivariate time series classification with hierarchical variational graph pooling (2020). arXiv preprint arXiv:2010.05649
Zhang, X., et al.: TapNet: multivariate time series classification with attentional prototypical network. In: Proceedings of the AAAI Conference on Artificial Intelligence, Apr 2020, vol. 34, no. 04, pp. 6845–6852. https://doi.org/10.1609/aaai.v34i04.6165
Xiao, Z., et al.: RTFN: a robust temporal feature network for time series classification. Inf. Sci. (2020). https://doi.org/10.1016/j.ins.2021.04.053
Lei, Y., Wu, Z.: Time series classification based on statistical features. EURASIP J. Wireless Commun. Netw. (2020). https://doi.org/10.1186/s13638-020-1661-4
Tripathi, A.M., et al.: Multivariate time series classification with an attention-based multivariate convolutional neural network. In: 2020 International Joint Conference on Neural Networks (IJCNN), July 2020
Lin, H., et al.: A multivariate time series classification method based on self-attention. Genet. Evol. Comput. (2020). https://doi.org/10.1007/978-981-15-3308-254
Karima, F., et al.: Multivariate LSTM-FCNs for time series classification. arXiv:1801.04503 [cs.LG], July 2019. https://doi.org/10.1016/j.neunet.2019.04.014
Schafer, P., et al.: Multivariate time series classification with WEASEL+MUSE. arXiv:1711.11343v4 [cs.LG], Aug 2018. https://doi.org/10.1145/nnnnnnn.nnnnnnn
Helwig, N., et al.: Condition monitoring of a complex hydraulic system using multivariate statistics. In: 2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings, July 2015. https://doi.org/10.1109/I2MTC.2015.7151267
Helwig, N., et al.: Detecting and compensating sensor faults in a hydraulic condition monitoring system. In: SENSOR 2015, Nuremberg, May 2015. https://doi.org/10.5162/sensor2015/D8.1
Christ, M., et al.: Time series feature extraction on basis of scalable hypothesis tests (tsfresh a python package). Neurocomputing (2018). https://doi.org/10.1016/j.neucom.2018.03.067
Barandas, M., et al.: TSFEL: Time Series Feature Extraction Library, 100456 (2020). ISSN 2352-7110. https://doi.org/10.1016/j.softx.2020.100456
Schowe, B.: Feature Selection for High-Dimensional Data with RapidMiner. Technical University of Dortmund (2010)
Ratanamahatana, C.A., Lin, J., Gunopulos, D., Keogh, E., Vlachos, M., Das, G.: Data Mining and Knowledge Discovery Handbook, 2nd edn. In: Maimon, O., Rokach, L. (eds.), pp. 1049–1077. Springer (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Sirisha, B., Goli, S.G., Balram, J., Manoj, A.V.S.S., Praneeth, R., Sandhya, B. (2022). A Featurized Learning Approach for Health Monitoring in Hydraulic Systems Using Multivariate Time Series Data. In: Satapathy, S.C., Bhateja, V., Favorskaya, M.N., Adilakshmi, T. (eds) Smart Intelligent Computing and Applications, Volume 2. Smart Innovation, Systems and Technologies, vol 283. Springer, Singapore. https://doi.org/10.1007/978-981-16-9705-0_7
Download citation
DOI: https://doi.org/10.1007/978-981-16-9705-0_7
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-9704-3
Online ISBN: 978-981-16-9705-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)