Fault Detection of Aerospace Systems Based on Binary Grasshopper Optimization Algorithm

Abstract

The entire flight movement procedure is managed by the aerospace system.Its ability to detect faults can help the aerospace prognostic health management system make decisions and carry out targeted maintenance, which is crucial for enhancing the safety and dependability of the air- craft systems. Aerospace systems are constantly subject to several failures due to the risks and difficulties of the space environment, including the deterioration of subsystem performance, sensor errors, connection loss, or equipment damage.The fault diagnosis method for aerospace systems based on binary grasshopper optimisation algorithm is proposed in this study using Deep Learning (DL) technique by taking use of the strong learning and intelligent recognition capacity.The suggested system offered a novel LSTM autoencoder architecture with supervised machine learning and deep learning techniques to carry out two distinct stages of fault diagnosis. The detection phase, employing the LSTM autoencoder with KNN, compacted the two phases. Then, the fault diagnosis phase, which is represented by the classification schema, is updated using a decision tree with KNN.The fault detection and diagnostics for LSTM in aircraft systems was completed successfully. The experimental findings proved the superiority and efficacy of the suggested strategy. The experimental findings proved the efficacy and superiority of the suggested strategy.