Fatigue Crack Growth Life Prediction of 6061 Al-Alloy under Load Ratio Effect by Using ANFIS

J. R. Mohanty

Adaptive neuro-fuzzy inference system (ANFIS); fatigue crack growth rate (FCGR); root mean squre error (RMSE); prediction ratio.

Fatigue crack growth under constant amplitude loading for a particular material strongly depends on load ratio (R). The prediction of fatigue crack growth life under such situation through deterministic approach is a tedious task. Application of artificial intelligence methods is more encouraging in those complex situations. In the present work a novel soft-computing approach i.e. adaptive neuro-fuzzy technique (ANFIS) has been applied to predict fatigue life of 6061 (AA 6061) aluminum alloy under the influence of load ratio. It has been observed that the ANFIS model predict the fatigue life of the alloy reasonably well with percentage deviation of –0.024 and prediction ratio of 1.025.

10.22161/ijaers/3.11.30