Hardness Prediction Model for En Grade Steels Subjected to Different Heat Treatment Processes

A B S T R A C T It is more often witnessed that the hardness of the steel depends on environment conditions, type of heat treatment adopted, composition and morphology. The selection of process parameters plays a vital role in obtaining the required hardness. It opens up scope for extensive research to map the relationship between the process parameters which is coherent with hardness of the steel. In the present investigation, an attempt has been made to accomplish this task with the help of a support vector machines (SVM) model for mapping process parameters with hardness. The basis for the development of SVM prediction model for the hardness at any condition within the conducted domain has been obtained by the data base comprising of set of input variable such as process, temperature, metal grade and output variable such as hardness. This is achieved by conducting several experimentations at different temperatures for various heat treatment processes such as annealing, normalizing, hardening and quenching using two different grades of steels namely EN19 and EN24 (with and without nickel). The presence of Nickel, which is an austenite stabilizer, promotes the formation of needle like fine grain martensite phase and its effect on hardness has been reported.