Using GMM-HMM Model and Parallel Computing for Health Estimation and Prognosis of Turbofan Engines
- 10.2991/cmsa-18.2018.15How to use a DOI?
- GMM; HMM; Aero-enigine; RUL
Estimating the residual useful life (RUL) time of turbofan engines is a determinant factor for aviation safety. Different methods of machine learning are explored in the field of turbofan health estimation and prognosis, to save maintenance cost and enhance aviation safety. An algorithm used for turbofan engines residual user life estimation called gaussian mixture model and hidden Markov model is one of the most effective method many researchers used in complex system health estimation and prognosis. However, Gaussian Mixture Model and Hidden Markov Model (GMM-HMM) has N2 computational complexity, which is always time consuming under the condition of large dataset is available. During past two decades, along with more and more advanced sensors are used for turbofan engine health monitoring, since a large amount of data is available for observation and analyzation. Parallel computing library such as CUDA is extremely necessary. In this paper, CUDA library is adopted for turbofan engines RUL estimation. The performance of algorithms is demonstrated on a simulated aviation dataset generated by the Commercial Modular Aero-Propulsion System Simulation (CMAPSS).
- © 2018, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Zilu Wang PY - 2018/04 DA - 2018/04 TI - Using GMM-HMM Model and Parallel Computing for Health Estimation and Prognosis of Turbofan Engines BT - Proceedings of the 2018 International Conference on Computer Modeling, Simulation and Algorithm (CMSA 2018) PB - Atlantis Press SP - 63 EP - 67 SN - 1951-6851 UR - https://doi.org/10.2991/cmsa-18.2018.15 DO - 10.2991/cmsa-18.2018.15 ID - Wang2018/04 ER -