Fault Detection Method for Aircraft Air Conditioning Heat Exchanger Based on SDAE-SMC

The aircraft air conditioning system is one of the important onboard equipment in modern civil aircraft. By circulating the air inside and outside the cabin, it ensures suitable pressure, temperature, and air quality inside the cabin. However, faults often occur in the air conditioning system, especially in its pressurization and air conditioning components, among which the problem of poor cooling effect is particularly prominent due to contamination and blockage of the heat exchanger. A fault detection method based on stacked denoising autoencoder and Softmax classifier is proposed in this paper. Firstly, the heat dissipation efficiency of the heat exchanger is derived based on the principles of fluid thermodynamics. Deep feature mining is performed using SDAE to extract deep features from high-dimensional sensor data, and SMC is used to classify these features, achieving fault detection of the air conditioning heat exchanger. The experimental results show that this method can effectively distinguish between faulty samples and healthy samples, and the fault detection results exhibit excellent performance in accuracy, recall, and F1 score, which is significantly improved compared to traditional methods.