Fatigue crack growth prediction for spectrum loadings using neural networks

Conference · Thu Dec 31 23:00:00 EST 1992

OSTI ID:6051366

Pidaparti, R M.V.; Palakal, M

An artificial neural network method is developed to represent the fatigue crack growth and cycle relationships under different spectrum loadings. The method utilizes load cycle spectrum using available flight data and experimental data for crack growth vs cycles as input. The trained network is able to predict the relationship between the crack growth and loading cycles. The neural network is able to generalize the crack growth-cycle behavior for different variations in the loading spectrums. The result predicted by the neural network model seems reasonable and the model is capable of representing crack growth behavior for arbitrary loadings. 7 refs.

• OSTI ID: 6051366
• Report Number(s): AIAA-Paper--93-1609; CONF-9304186--
• Country of Publication: United States
• Language: English

Related Subjects

36 MATERIALS SCIENCE
360103* -- Metals & Alloys-- Mechanical Properties
99 GENERAL AND MISCELLANEOUS
990200 -- Mathematics & Computers
ARTIFICIAL INTELLIGENCE
CRACK PROPAGATION
CRACKS
DYNAMIC LOADS
FATIGUE
FORECASTING
KNOWLEDGE BASE
MECHANICAL PROPERTIES
NEURAL NETWORKS