Prediction of fracture toughness temperature dependence applying neural network

0366644 - ÚFM 2012 RIV RS eng J - Článek v odborném periodiku
Dlouhý, Ivo - Hadraba, Hynek - Chlup, Zdeněk - Šmída, T.
Prediction of fracture toughness temperature dependence applying neural network.
Structural Integrity and Life. Roč. 11, č. 1 (2011), s. 9-14. ISSN 1451-3749
Grant CEP: GA ČR(CZ) GAP108/10/0466
Výzkumný záměr: CEZ:AV0Z20410507
Klíčová slova: brittle to ductile transition * fracture toughness * artificial neural network * steels
Kód oboru RIV: JL - Únava materiálu a lomová mechanika

Reference temperature localizing the fracture toughness temperature diagram on temperature axis is predicted based on tensile test data. The regularization neural network is developed to solve the correlation of these properties. Three-point bend specimens were applied to determine fracture toughness. The fracture toughness transition dependence is quantified by means of master curve concept enabling to represent it by using one parameter, i.e. reference temperature. Tensile samples with circumferential notch are also examined. In total 29 data sets from low-alloy steels are applied for the analysis. A good correlation of predicted and experimentally determined values of reference temperature is found.
Trvalý link: http://hdl.handle.net/11104/0201552