DocumentCode :
3241836
Title :
Prediction of fatigue crack growth rate using rule-based systems
Author :
AbdulRazzaq, M. ; Ariffin, A.K. ; El-Shafie, A. ; Abdullah, S. ; Sajuri, Z.
Author_Institution :
Dept. of Mech. & Mater. Eng., Nat. Univ. of Malaysia (UKM), Bangi, Malaysia
fYear :
2011
fDate :
19-21 April 2011
Firstpage :
1
Lastpage :
8
Abstract :
Fatigue crack propagation life has been estimated based on established empirical equations. In the present work, fatigue crack rates of ASTM A533 alloy under the different load histories was predicted by adaptive neuro fuzzy approach (ANFIS). A novel soft-computing application, suitable for non-linear, noisy and complex problems like fatigue. The features and usefulness of our method are to initial the fatigue crack growth rate with the number of cycle´s relationship for each case study. In this research, an automatic prediction methodology has been adopted to estimate the constant amplitude loading fatigue life under the above condition by applying neuro fuzzy approach (ANFIS). Soft-computing methods show great potential for predicting the fatigue crack growth rate, especially with really data. The predicted results are found to be in perfect agreement with the experimental findings when tested on ASTM A533 alloy.
Keywords :
fatigue cracks; fuzzy neural nets; fuzzy reasoning; knowledge based systems; mechanical engineering computing; nonlinear programming; ANFIS; ASTM A533 alloy; adaptive neuro fuzzy approach; automatic prediction methodology; constant amplitude loading fatigue life; empirical equations; fatigue crack growth rate; fatigue crack propagation life; fatigue crack rates; load history; noisy and complex problems; nonlinear problems; rule-based systems; soft-computing application; Adaptation model; Artificial neural networks; Fatigue; Load modeling; Loading; Mathematical model; Training; Constant amplitude loading; Fatigue crack growth rate; Load sequence; adaptive neuro-fuzzy inference system (ANFIS);
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Modeling, Simulation and Applied Optimization (ICMSAO), 2011 4th International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4577-0003-3
Type :
conf
DOI :
10.1109/ICMSAO.2011.5775513
Filename :
5775513
Link To Document :
بازگشت