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.
The various interfering signals present in the space result in the inability to obtain a clean and clear solar radio dynamic spectrum, which affects the effective observation of solar radio burst events. At the data processing level, we propose a method for predicting and processing radio interference signals in solar radio burst events using recurrent neural network in deep learning. Firstly, the radio signal that satisfies the condition is selected by the amount of solar radio flux of all frequency channels at some moment, and then the position of the initial time of the burst event is located by using the variation curve of the solar radio flux with time under the single frequency channel. After that, the constructed recurrent neural network is used to predict the signal value of the radio in the burst area. Finally, according to the linear additivity of the signal, the value of the clean pure burst event is obtained by subtracting the predicted radio value from the original value of the burst area. The experimental results show that the proposed method can effectively remove the interference in the solar radio dynamic spectrum and preserve the effective information of the burst event to the greatest extent. This provides a new idea and research direction for deep learning in the anti-interference processing of astronomical big data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.