Font Size: a A A

Research On Prediction Of Remaining Useful Life Of DA40 Aircraft Brake Pads Based On Long Short-term Memory Network

Posted on:2022-01-03Degree:MasterType:Thesis
Country:ChinaCandidate:Y K WangFull Text:PDF
GTID:2532306488980429Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
Aircraft brake pad is a key safety part in aircraft brake system,and its service performance directly affects the reliability and stability of aircraft operation.Accurate remaining useful life prediction of brake pads is of great theoretical value and practical significance for accurately evaluating the working condition of brake pads,improving the safety during the use of brake pads and reducing the occurrence of major braking accidents.In this paper,the brake pad of airplane DA40 is taken as the research object.Based on deep learning method and time series signal decomposition technology,the remaining life prediction of brake pad is studied.Based on the strong non-linear and non-stationary characteristics of the wear degradation process of aircraft brake pads,a method for predicting the remaining useful life of aircraft brake pads based on Variable Mode Decomposition and Bidirectional Long Short-Term Memory network is proposed.Firstly,the VMD algorithm is used to decompose the brake pad wear data into several relatively stable subsequence components,then a BiLSTM prediction model is built for each subsequence,finally,the prediction results of each subsequence are added to get the remaining useful life prediction results of aircraft brake pads.In order to further improve the prediction effect of the model,the high-frequency residual term of VMD is decomposed by the Ensemble Empirical Mode Decomposition(EEMD),and the decomposed subsequence is reconstructed by the sample entropy method to extract the eigenvector,and a BiLSTM prediction model is constructed for prediction,which is superimposed with the prediction result of the VMD-BiLSTM model as the final RUL prediction result of the aircraft brake pads.Finally,the wear data and operating condition data of the brake pads of DA40 aircraft are used as the data set for experiments and compared with several prediction models.The results show that the proposed method can effectively improve the accuracy of aircraft brake pad remaining useful life prediction,and provide effective maintenance strategies for the maintenance of equipment.
Keywords/Search Tags:Aircraft brake pads, Remaining Useful Life prediction, Deep learning, Bidirectional Long Short-Term Memory, Variational Mode Decomposition
PDF Full Text Request
Related items