Model of a Predictive Neural Network for Determining the Electric Fields of Training Flight Phases

Abstract

Overview

Electromagnetic fields (EMFs) in aviation training environments are a significant concern, due to potential safety risks to crew and passengers. This study introduces an innovative neural network aimed at measuring and predicting EMF intensities during training flights.

Methodology

• A series of EMF measurements was performed using the NHT3DL broadband meter from Microrad, equipped with a 01E measuring probe, covering frequencies from 100 kHz to 6.5 GHz.
• Data was collected from flights on several aircraft models including Cessna C172, Cessna C152, Aero AT3, and Technam P2006T.
• The neural network utilized LSTM layers to effectively predict average and instantaneous EMF values encountered during flights.

Findings

The artificial neural network demonstrated higher accuracy compared to traditional methods in predicting EMF exposure, enhancing the potential for optimizing flight routes to minimize EMF risks.

Conclusion

The study underscores the importance of continuous EMF monitoring and the application of predictive technologies to safeguard the health of flight crews and passengers by adhering to normative EMF limits. This approach not only contributes to EMF safety but also promotes electromagnetic compatibility and environmental conservation in general aviation.