Davut, KemalTuzun, Mert YagizYalcin, Mustafa AlpDavut, KemalKilicli, VolkanDepartment of Metallurgical and Materials Engineering2024-07-052024-07-05202310025-53002195-857210.1515/mt-2022-02652-s2.0-85149890272https://doi.org/10.1515/mt-2022-0265https://hdl.handle.net/20.500.14411/2489Kilicli, Volkan/0000-0002-0456-5987; Yalcin, Mustafa Alp/0000-0001-5371-4574Austempered ductile iron (ADI) has been preferred in a wide range of applications due its unique combination of high strength, good ductility, wear resistance and fracture toughness together with lower cost and lower density compared to steels. Magnetic Barkhausen noise (MBN) measurement offers a better alternative to traditional characterization techniques by being fast and non-destructive. A simple linear regression using only one single independent variable cannot correlate the MBN with the microstructure of ADI, since its microstructure is multi component. Multiple linear regression analysis (MLRA) was used to build a model that uses the characteristic features of microstructural constituents as input parameters to predict the MBN. For that purpose, Cu-Ni-Mo alloyed ductile iron samples austempered between 325 and 400 degrees C and for 45-180 min duration were used. The results show that MBN is most sensitive to the size and shape of acicular ferrite and retained austenite. Moreover, MBN is almost insensitive to the size, morphology and volume fraction of graphite particles. This indicates that retained austenite pins the domain walls more effectively than the graphite particles. Considering the results MLRA, MBN technique can be used to characterize the ausferritic microstructure of ADI.eninfo:eu-repo/semantics/closedAccessaustempered ductile iron (ADI)magnetic Barkhausen noise (MBN)microstructuremultiple linear regression analysis (MLRA)retained austeniteArticle653453465WOS:000945936400014