Abstract
The accurate prediction of airplane engine failure can provide a reasonable decision basis for airplane engine maintenance, effectively reducing maintenance costs and reducing the incidence of failure. According to the characteristics of the monitoring data of airplane engine sensors, this work proposed a remaining useful life (RUL) prediction model based on principal component analysis and bidirectional long short-term memory. Principal component analysis is used for feature extraction to remove useless information and noise. After this, bidirectional long short-term memory is used to learn the relationship between the state monitoring data and remaining useful life. This work includes data preprocessing, the construction of a hybrid model, the use of the NASA’s Commercial Aerodynamic System Simulation (C-MAPSS) data set for training and testing, and the comparison of results with those of support vector regression, long short-term memory and bidirectional long short-term memory models. The hybrid model shows better prediction accuracy and performance, which can provide a basis for formulating a reasonable airplane engine health management plan.
Highlights
With the development of an airplane engine towards integration, systematization and precision, the engine system becomes more and more complex
The hybrid model was adopted to conduct remaining useful life (RUL) prediction, and principal component analysis (PCA) was used as a dimensionality reduction tool to extract the characteristics of the status monitoring data of the airplane engine
In the research on the remaining useful life of airplane engines, the evaluation criteria are mainly the root mean square error (RMSE) and the asymmetric scoring function to quantitatively evaluate the performance of models [21]
Summary
With the development of an airplane engine towards integration, systematization and precision, the engine system becomes more and more complex. The data-driven method refers to the use of such methods as the neural network to extract the features and the relationship between the data from the state monitoring data of the equipment, so as to achieve the prediction of the equipment’s RUL This is the application of artificial intelligence technology in fault prediction, which has more advantages than other methods. The feature extraction method of this work should achieve three objectives: removing noise and redundant information, improving the accuracy of the model; reducing the computational complexity of the model, and improving the calculation efficiency; ensuring that the reduced dimension data can retain most of the useful information in the original data. The characteristic data are input into the BLSTM, and the advantage of BLSTM in processing long time series and the characteristics of forward and backward two-way propagation are utilized to obtain the relationship between the status monitoring data and RUL, so as to obtain more accurate prediction results
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
