Hostname: page-component-848d4c4894-r5zm4 Total loading time: 0 Render date: 2024-06-21T21:22:08.422Z Has data issue: false hasContentIssue false
  • English
  • Français

Ensemble approaches for leveraging machine learning models in load estimation

Part of: The 19th and 20th Australian International Aerospace Congresses

Published online by Cambridge University Press:  03 November 2023

C. Cheung Open the ORCID record for C. Cheung [Opens in a new window] ,
E. Seabrook Open the ORCID record for E. Seabrook [Opens in a new window] ,
J.J. Valdés Open the ORCID record for J.J. Valdés [Opens in a new window] ,
Z.A. Hamaimou Open the ORCID record for Z.A. Hamaimou [Opens in a new window]  and
C. Biondic Open the ORCID record for C. Biondic [Opens in a new window]
C. Cheung*
Affiliation:
Aerospace Research Centre, National Research Council Canada, 1200 Montreal Rd, Ottawa, ON K1A 0R6, Canada
E. Seabrook
Affiliation:
Aerospace Research Centre, National Research Council Canada, 1200 Montreal Rd, Ottawa, ON K1A 0R6, Canada
J.J. Valdés
Affiliation:
Digital Technologies Research Centre, National Research Council Canada, 1200 Montreal Rd, Ottawa, ON K1A 0R6, Canada
Z.A. Hamaimou
Affiliation:
Aerospace Research Centre, National Research Council Canada, 1200 Montreal Rd, Ottawa, ON K1A 0R6, Canada
C. Biondic
Affiliation:
Aerospace Research Centre, National Research Council Canada, 1200 Montreal Rd, Ottawa, ON K1A 0R6, Canada
*
Corresponding author: C. Cheung; Email: catherine.cheung@nrc-cnrc.gc.ca
Get access Rights & Permissions [Opens in a new window]

Abstract

Helicopter component load estimation can be achieved through a variety of machine learning techniques and algorithms. A range of ensemble integration techniques were investigated in order to leverage multiple machine learning models to estimate main rotor yoke loads from flight state and control system parameters. The techniques included simple averaging, weighted averaging and forward selection. Performance of the models was evaluated using four metrics: root mean squared error, correlation coefficient and the interquartile ranges of these two metrics. When compared, every ensemble outperformed the best individual model. The ensembles using forward selection achieved the best performance. The resulting output is more robust, more highly correlated and achieves lower error values as compared to the top individual models. While individual model outputs can vary significantly, confidence in their results can be greatly increased through the use of a diverse set of models and ensemble techniques.

Keywords

load estimationhealth and usage monitoringintegrated vehicle health managementensemblesmachine learning
Type
Research Article
Information
The Aeronautical Journal , Volume 127 , Special Issue 1318: AIAC 2023 , December 2023 , pp. 2082 - 2104
Check for updates
Copyright
© National Research Council Canada, 2023. Published by Cambridge University Press on behalf of Royal Aeronautical Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

A version of this paper first appeared at The Australian International Aerospace Congress 2021 (AIAC19).

References

Cheung, C., Sehgal, S. and Valdés, J.J. A machine learning approach to load tracking and usage monitoring for legacy fleets. ICAF 2019 – Structural Integrity in the Age of Additive Manufacturing: Proceedings of the 30th Symposium of the International Committee on Aeronautical Fatigue, June 2–7, A. Niepokolczycki and J. Komorowski (eds), Krakow, Poland, Springer Cham, 2019, pp 922–937. doi: https://doi.org/10.1007/978-3-030-21503-3.CrossRefGoogle Scholar
Cheung, C., Rocha, B., Valdés, J.J. and Puthuparampil, J. Improved load estimation and fatigue life tracking demonstrated on multiple platforms using the signal approximation method. American Helicopter Society 72nd Annual Forum Proceedings, Fairfax, VA, 2016, pp 1–14.Google Scholar
Cheung, C., Biondic, C., Hamaimou, Z.A. and Valdés, J.J. An approach to merging machine learning models in an ensemble for load estimation. Proceedings of the 12th DST International Conference on Health and Usage Monitoring (HUMS2021), Defence Science and Technology Group, Melbourne, Australia, 2021, pp 1–8.Google Scholar
Cheung, C. and Hamaimou, Z.A. Ensemble integration methods for load estimation. Vertical Flight Society 78th Annual Forum Proceedings, Fairfax, VA, 2022, pp 1–9.CrossRefGoogle Scholar
Polanco, F.G. Estimation of Structural Component Loads in Helicopters: A Review of Current Methodologies. Technical Note DSTO-TN-0239. Melbourne, VIC: DSTO Aeronautical and Maritime Research Laboratory, 1999, pp 1–31.Google Scholar
Oldersma, A. and Bos, M.J. Airframe loads and usage monitoring of the CH-47D “Chinook” helicopter of the Royal Netherlands Air Force. In ICAF 2011 Structural Integrity: Influence of Efficiency and Green Imperatives (Komorowski, J., ed.), pp 473–493. Dordrecht: Springer Netherlands, 2011.Google Scholar
Qing, X., Liao, Y., Wang, Y., Chen, B., Zhang, F. and Wang, Y. Machine learning based quantitative damage monitoring of composite structure. Int. J. Smart Nano Mater., 2022, 13, (2), pp 167–202.CrossRefGoogle Scholar
Mucha, W. Comparison of machine learning algorithms for structure state prediction in operational load monitoring. Sensors, 2020, 20, (24), p 7087.CrossRefGoogle ScholarPubMed
Del cistia Gallimard, C., Beroul, F., Denoulet, J., Nikolajevic, K., Pinna, A., Granado, B. and Marsala, C. Harmonic decomposition to estimate periodic signals using machine learning algorithms: Application to helicopter loads. 2022 International Joint Conference on Neural Networks (IJCNN), IEEE, Padua, Italy, pp 1–8, 2022.Google Scholar
Isom, J., Fletcher, J., Davis, M., Cycon, J. and Rozak, J. Flight test of technology for virtual monitoring of loads. American Helicopter Society 69th Annual Forum Proceedings, Phoenix, AZ, 2013, pp 1–20.Google Scholar
Raymond B. Jr., Davis, M., Kloda, J. and Templeton, D. S-92 main rotor blade life extension. American Helicopter Society 72nd Annual Forum Proceedings, West Palm Beach, FL, USA, May 2016, pp 1–10.Google Scholar
Owen, I., White, M.D., Padfield, G.D. and Hodge, S.J. A virtual engineering approach to the ship-helicopter dynamic interface – A decade of modelling and simulation research at the University of Liverpool. Aeronaut. J., 2017, 121, (1246), pp 1833–1857.CrossRefGoogle Scholar
Ling, Y. and Mahadevan, S. Integration of structural health monitoring and fatigue damage prognosis. Mech. Syst. Signal Process., 2012, 28, pp 89–104. Interdisciplinary and Integration Aspects in Structural Health Monitoring.CrossRefGoogle Scholar
Cheung, C., Rocha, B., Valdés, J.J., Kotwicz-Herniczek, M. and Stefani, A. Expanded fatigue damage and load time signal estimation for dynamic helicopter components using computational intelligence techniques. American Helicopter Society 70th Annual Forum Proceedings, Montréal, Québec, Canada, May 2014, pp 1–10.Google Scholar
Friedman, J.H. Multivariate adaptive regression splines. Ann. Statist., 1991, 19, (1), pp 1–67.Google Scholar
Breiman, L. Random forests. Mach. Learn., 2001, 45, (1), pp 5–32. doi: https://doi.org/10.1023/A:1010933404324.Google Scholar
Hochreiter, S. and Schmidhuber, J. Long short-term memory. Neural Comput., 1997, 9, pp 1735–1780.CrossRefGoogle ScholarPubMed
Pal, S. and Mitra, S. Multilayer perceptron, fuzzy sets, and classification. IEEE Trans. Neural Netw., 1992, 3, (5), pp 683–697.CrossRefGoogle ScholarPubMed
Lecun Y., Bottou L., Bengio Y. and Haffner P. Gradient-based learning applied to document recognition, in Proceedings of the IEEE, 86, (11), pp. 2278–2324, Nov. 1998, doi: https://doi.org/10.1109/5.726791.CrossRefGoogle Scholar
Dietterich, T.G. Machine-learning research. AI Mag., 1997, 18, p 97.Google Scholar
Webb, G. and Zheng, Z. Multistrategy ensemble learning: reducing error by combining ensemble learning techniques. IEEE Trans. Knowl. Data Eng., 2004, 16, (8), pp 980–991.CrossRefGoogle Scholar
Mendes-Moreira, J.A., Soares, C., Jorge, A.M. and De Sousa, J.F. Ensemble approaches for regression: A survey. ACM Comput. Surv., 2012, 45, (1), Article 10 (pages 1–40). doi: https://doi.org/10.1145/2379776.2379786.CrossRefGoogle Scholar
Breitling, R., Armengaud, P., Amtmann, A. and Herzyk, P. Rank products: A simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments. FEBS Lett., 2004, 573, (1), pp 83–92.CrossRefGoogle ScholarPubMed
Tsoumakas, G., Partalas, I. and Vlahavas, I. A taxonomy and short review of ensemble selection. Proceedings ECAI'2008 Workshop on supervised and unsupervised ensemble methods and their applications (SUEMA'2008), 2008, pp 41–46.Google Scholar
Dietterich, T.G. Ensemble methods in machine learning. Multiple Classifier Systems, Berlin, Heidelberg, Springer Berlin Heidelberg, pp 1–15, 2000.CrossRefGoogle Scholar
Martnez-Muñoz, G. and Suárez, A. Pruning in ordered bagging ensembles. Proceedings of the 23rd International Conference on Machine Learning, ICML ’06, New York, NY, USA, Association for Computing Machinery, pp 609–616, 2006.CrossRefGoogle Scholar