The aerospace industry is on the brink of a significant transformation in how it approaches the safety and reliability of composite structures. Recent research led by Ferda C. Gül from the Center of Excellence in Artificial Intelligence for Structures at Delft University of Technology introduces an integrated methodology to predict the Remaining Useful Life (RUL) of these structures under in-plane compressive fatigue loading. This innovation could have far-reaching implications not just for aerospace but also for the construction sector, where composite materials are increasingly being used.
Composite structures are known for their strength and lightweight properties, making them ideal for various applications. However, their complex degradation behavior poses a challenge for maintaining structural integrity over time. Gül’s study sheds light on this issue by utilizing a Deep Neural Network (DNN) trained with Guided Wave-based damage indicators (GW-DIs). These indicators are derived through advanced signal processing techniques, including Hilbert transform and Continuous Wavelet Transform, which can detect damage accumulation that could lead to failure.
“The ability to monitor damage mechanisms in real-time is a game changer,” Gül stated. “Our methodology not only predicts RUL but also tracks the progression of delamination, which is a critical factor in the longevity of composite structures.” This proactive approach allows for timely interventions, ultimately enhancing safety and reducing costs associated with unexpected failures.
The research employs two distinct frameworks: one that analyzes individual samples to understand path dependency in RUL and delamination prognosis, and another that uses an ensemble dataset to create a more generalized model applicable across various stress conditions. This dual approach enables the model to encapsulate both fast and slow degradation scenarios, offering a nuanced understanding of how composite materials behave under stress.
The implications for the construction industry are profound. As projects increasingly incorporate composite materials for their durability and efficiency, the ability to predict RUL accurately can lead to significant cost savings and improved safety standards. Structures can be monitored more effectively, and maintenance schedules can be optimized, reducing downtime and extending the lifecycle of critical infrastructure.
This research, published in ‘Composites Part C: Open Access’ (translated to English as ‘Composites Part C: Open Access’), paves the way for future developments in structural health monitoring and predictive maintenance. With advancements in artificial intelligence and machine learning, the construction sector can look forward to a future where safety and efficiency go hand in hand.
For more information on this groundbreaking work, you can visit the Center of Excellence in Artificial Intelligence for Structures.
