Volume 38, Number 3, June 2020
|Page(s)||507 - 514|
|Published online||06 August 2020|
Improved Model for On-Board Real-Time by Constructing Empirical Model via GMM Clustering Method
School of Power and Energy, Northwestern Polytechnical University, Xi'an 710072, China
The method of constructing an empirical model is used to compensate the deviation between the output of the on-board real-time model and the engine measurement parameters, and improve the parameter tracking and estimation performance of the on-board adaptive model in the full flight envelope. Due to the large amount of data acquired online, the clustering method based on Gaussian mixture model is implemented to realize data compression for offline training and updating the empirical model. The present empirical model is applied to the on-board adaptive model of civil large bypass ratio turbofan engine. The simulation results show that the empirical model based on Gaussian mixture model can reduce the output error of on-board real-time model, and the accuracy of the health parameter estimation and engine component fault isolation performance of the on-board real-time adaptive model with empirical model are improved.
Key words: aero-engine / onboard real-time model / gaussian mixture model / empirical model
关键字 : 航空发动机 / 机载实时模型 / 高斯混合模型 / 经验模型
© 2019 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.