Operational Load Monitoring of a Composite Panel Using Artificial Neural Networks

Area of Science:

• Aerospace Engineering
• Structural Health Monitoring
• Computational Mechanics

Background:

• Operational Load Monitoring is crucial for predicting the remaining service life of structures, especially in aerospace.
• Traditional methods require numerous sensors, increasing complexity and cost.
• Variable exciting forces necessitate comprehensive data acquisition.

Purpose of the Study:

• To propose an artificial neural network (ANN) approach for estimating structural stress states using limited sensor data.
• To reduce the number of required sensors in operational load monitoring.
• To enhance the reliability of lifetime predictions for aerospace components.

Main Methods:

• Training an ANN using finite element simulations of an omega-stiffened composite panel.
• Acquiring data from a limited number of strain gauges (six in the example).
• Validating the ANN model with experimental data under various load cases.

Main Results:

• The trained ANN accurately estimated the structural state based on data from a few sensors.
• Experimental validation confirmed the model's predictive capabilities.
• The approach demonstrated the feasibility of reducing sensor count without compromising accuracy.

Conclusions:

• Artificial neural networks offer a viable solution for efficient and reliable operational load monitoring.
• This method can extend component lifespan in aerospace without compromising flight safety.
• The study validates the ANN approach for composite structural panels.