A multi-region nonlinear model for fatigue life prediction of 3D woven composites under combined tension and bending fatigue

Aero-engine fan blades are the beating hearts of modern aircraft, spinning at incredible speeds to generate the thrust that lifts planes into the sky. To make these engines more efficient and lighter, engineers are increasingly turning to a special class of materials called three-dimensional woven composites. Unlike traditional materials that are layered like sheets of paper, these composites are built like a complex, interlocking fabric where fibers weave through the entire thickness of the part, creating a structure that is both incredibly strong and resistant to coming apart. However, these blades do not just spin in a steady rhythm; they endure a chaotic mix of forces. As the engine spins, the blades face a heavy, slow-pulling force from centrifugal power, while simultaneously being battered by rapid, high-frequency vibrations from air turbulence and mechanical noise. This combination of slow, heavy stress and fast, light shaking creates a unique type of wear and tear that is difficult to predict. If engineers cannot accurately forecast when these blades will fail, the consequences could be catastrophic for flight safety.

For decades, scientists have tried to predict how long materials last under such mixed conditions using simple rules that assume damage adds up in a straight, predictable line. The idea was that a little bit of damage from the slow pull and a little bit from the fast shake would simply sum together to reach a breaking point. But real-world materials, especially these complex woven fabrics, do not behave so neatly. The interaction between the slow pull and the fast shake is messy and nonlinear, meaning the damage they cause together is often far worse than the sum of their parts. Researchers at the Nanjing University of Aeronautics and Astronautics set out to solve this puzzle by developing a new way to model how these materials degrade. They recognized that the damage process is not uniform; instead, it shifts through different stages depending on how intense the vibrations are relative to the pulling force.

To understand these shifts, the team first built a specialized testing machine capable of applying both types of stress at the same time. They designed a unique fixture that could hold a sample of the woven composite under a heavy tensile load while simultaneously shaking it side-to-side with high-frequency vibrations, ensuring the two forces did not interfere with each other. They tested two types of samples: solid blocks of the material and blocks with a hole in the center, which mimics the stress concentrations found where blades attach to the engine hub. By running these samples until they broke under various combinations of stress and vibration, the researchers gathered a rich set of data showing exactly how long the materials lasted under different conditions.

The data revealed that the relationship between the vibration intensity and the material's lifespan is not a simple curve. Instead, the researchers identified three distinct zones of behavior. In the first zone, where the vibrations are very weak compared to the pulling force, the vibrations have almost no effect, and the material fails mostly due to the heavy pull. In the second zone, as the vibrations get stronger, the material enters a "fuzzy" state where the damage becomes uncertain and harder to predict; the vibrations begin to interact with the pull in a way that accelerates wear, but the exact point of failure becomes blurry. Finally, in the third zone, where the vibrations are strong enough, the interaction becomes fierce, causing the material to fail much faster than expected. The key discovery was that there are specific thresholds where the material's behavior flips from one of these zones to another.

To capture this complexity, the researchers created a new mathematical model that treats these three zones differently. For the weak vibration zone, they used a straightforward calculation. For the strong vibration zone, they used a formula that accounts for the rapid acceleration of damage. But for the tricky middle zone, where the outcome is uncertain, they employed a concept from fuzzy logic. This approach allows the model to acknowledge that the boundary between "safe" and "dangerous" is not a sharp line but a gradual transition, much like the way twilight is neither fully day nor fully night. By using this fuzzy logic, the model could assign a degree of probability to the damage occurring in this transition zone, rather than forcing a rigid yes-or-no answer.

When the team tested their new model against the actual experimental data, it proved to be significantly more accurate than the traditional methods used in engineering today. The old methods tended to be overly cautious, predicting that the blades would fail much sooner than they actually did, which could lead to unnecessary and costly replacements. Other existing models were sometimes too optimistic, missing the danger entirely. The new multi-region model, however, predicted the lifespan of the blades with a high degree of precision, keeping its errors within a very narrow margin. It successfully accounted for how the material behaves under different stress levels and vibration frequencies, providing a reliable tool for engineers to design safer, longer-lasting fan blades.

The study also shed light on how the material actually breaks down at a microscopic level. In the weak vibration zone, the damage is isolated, with tiny cracks forming in the resin that holds the fibers together but not spreading far. As the vibrations increase into the fuzzy zone, these cracks begin to connect, and the fibers start to pull away from the resin in more places, weakening the structure's ability to transfer load. Once the vibrations cross into the strong zone, the damage becomes widespread, with large sections of the material losing their integrity as cracks spread rapidly through the layers. This detailed understanding of the damage evolution confirms that the new model is not just a statistical trick but reflects the real physical changes happening inside the material.

Ultimately, this research offers a clearer path forward for the design of advanced aircraft engines. By moving away from simple, linear assumptions and embracing the complex, shifting nature of how these materials fail, engineers can now predict fatigue life with greater confidence. This means they can push the boundaries of lightweight design without compromising safety, ensuring that the next generation of aircraft can fly more efficiently and reliably. The work demonstrates that even in the high-stakes world of aerospace engineering, acknowledging the uncertainty and complexity of the physical world leads to better, more accurate solutions.