Determination of S –N curves with the application of artificial neural networks P. ARTYMIAK, L. BUKOWSKI, J. FELIKS, S. NARBERHAUS 1 and H. ZENNER 1 Institute for Automation, TU Krakau, Poland, 1 Institute for Plant Engineering and Fatigue Analysis, TU Clausthal, Leibnizstrasse 32, D-38678 Clausthal-Zellerfeld, Germany Received in final form 22 April 1999 ABSTRACT The present paper describes the application of artificial neural networks for estimating the finite-life fatigue strength and fatigue limit. A comprehensive database with results of single-stage tests on specimens which simulate structural components is evaluated and prepared for processing with the use of neural networks. The available data are subdivided into different classes. A total of six different data classes are specified. The results of the prediction by means of neural networks are superior to those obtained with conventional methods for calculating the fatigue strength. The experimental results are estimated with high accuracy. Keywords S–N curves; artificial neural networks; constant amplitude tests NOMENCLATURE ANN=artificial neural network b =bias k =slope K t =notch factor n rel =relative number of load cycles N=number of load cycles N D =number of load cycles at transition point O=output P=input R=stress ratio R e =yield strength R m =tensile strength s=standard deviation s aD,rel =relative fatigue strength S aD =fatigue limit S ai =nominal stress amplitude T =scatter range W =weight W=potential Indices ANN =calculated with application of artificial neural networks calc =calculated results exp =experimental results synth =calculated by synthetic S–N curve method 10%=10% value 50%=mean value 90%=90% value © 1999 Blackwell Science Ltd. Fatigue Fract Engng Mater Struct 22, 723–728 723