Quantifying complex influences of chemical composition and soaking conditions for increasing the hot workability of M2 high-speed steel by using the alternative approach

Abstract

The conditions for increasing the hot workability and extending the temperature range for the safe hot working of M2 high-speed steel (HSS) were studied and revealed. This was enabled by combination of two approaches, i.e. results obtained by an analysis of so individual as well as spatial influences of chemical elements on the hot workability using a conditional average estimator neural networks in combination with the results obtained from hot-compression tests that revealed the appropriate soaking conditions. The Latin Hypercube Sampling technique was used to model the uncertainty of the collected data used in the analysis. The obtained results reveal new, surprisingly complex, typically spatial and (highly) non-linear relationships between the chemical elements and the hot workability of M2 HSS, i.e. common mutual influence of carbon, carbide-forming elements as well as elements, i.e. Si, Mn and Co, which indirectly influence the formation of carbides. Further also new allowed upper limits for contents of some harmful elements like S, P, Al, Sb, Cu, Sn, As, Ni, etc. at which transition from higher to lower workability takes place were revealed. Finally, by applying a specially developed procedure for hot-compression tests the appropriate soaking time and temperature were assessed. New findings explain and considerably improve the intrinsic hot workability and extend the temperature range for safe hot working at its upper and lower limits.