An objective is to improve existing life prediction models by incorporating  relevant material parameters and test conditions. To validate the performance of our technology, isothermal fatigue tests were performed on some of the TMC samples.

Our technique takes advantage of pattern recognition applied to Acoustic Emission (AE) inspection by transducers to detect stress waves. Numerous additional sources of acoustic emissions are available within composite materials due to their in-homogeneous nature and the corresponding  potential for failure at or near the matrix/reinforcement interfaces. Acoustic emission testing bestows tremendous potential to Neural Network technologies  for several reasons:

• The AE technique produces real-time feedback  that relates to the occurrence of micro structural damage in a composite  part. It has been demonstrated to be sensitive to several fracture modes  known to occur in fiber reinforced composites under cyclic loading.
• The AE technique is non-invasive. A variety of attachment methods may be utilized to affix  the transducer to the surface of a part, as dictated by the geometry and  temperature conditions of the component.
• AE can detect and evaluate the significance of discontinuities throughout an entire structure based upon readings taken  from a limited number of transducers. This allows AE to provide information  regarding the integrity of a structural component despite limited physical  access.
• AE is less sensitive to the geometry of  a component than most other non-destructive evaluation methods.