Deep learning detects electrical faults in aircraft power systems quickly
Deep Learning-Based Detection of Electrical Faults and Power Quality Disturbances in Aerospace Power Systems
Machine Learning
Summary
Electrical systems in modern aircraft operate at higher frequencies than normal power grids, making it harder to spot problems quickly. The authors created a realistic simulation of an airplane's electrical system that includes many types of faults and disturbances. They trained and tested deep learning models to detect these issues from electrical signals, with one model achieving nearly 97% accuracy. This model was also tested on specialized airplane hardware, running fast and accurately enough for real-time monitoring. This work shows that advanced AI can help keep airplane electrical systems safe by detecting faults promptly.
deep learningpower quality disturbanceelectrical fault detectionmulticlass classificationaerospace power systemshigh-frequency electrical networksconvolutional neural networktime-frequency analysismodel quantizationembedded edge AI
Authors
Ian C. Guzmán, Radu Babiceanu, Berker Peköz
Abstract
More Electric Aircraft require fast and reliable monitoring of high-frequency electrical networks, yet most power quality disturbance and fault diagnosis methods are developed for conventional 50 or 60 Hz grids. This work presents a hardware-aware deep learning framework for multiclass detection of electrical faults and power quality disturbances in a 400 Hz aerospace power system. A high-fidelity simulation model inspired by the Boeing 787 electrical architecture generates voltage and current waveforms for 21 normal, disturbance, switching, open-circuit, and short-circuit conditions. Two datasets, each containing 73,500 samples, are formed from one-dimensional time-series signals and short-time Fourier transform time-frequency representations. Signal-processing augmentation, domain randomization, and class-specific generative adversarial networks increase waveform diversity, and the time-series dataset is released through IEEE DataPort. We compare 1D and 2D convolutional neural networks, long short-term memory networks, CNN-LSTM hybrids, ResNet, MobileNet, and VGG models under common training conditions. A compact ResNet provides the best accuracy-complexity tradeoff, achieving 96.94 percent software test accuracy with 175,685 parameters. After 8-bit quantization and deployment on a Xilinx Zynq UltraScale Plus MPSoC ZCU102, the model achieves 95.87 percent accuracy and a measured mean neural-network accelerator latency of 6.90 ms per input record. The results establish simulation-based, accelerator-level feasibility for embedded edge AI in aircraft electrical health monitoring and motivate future end-to-end data acquisition and experimental validation.