Fuzzy rule-based prediction of hardness for sintered HSS components

In this work, an approach for predicting the hardness of sintered high speed steel (HSS) components using fuzzy rule-based model has been discussed. Sintered HSS components have been produced using powder metallurgy (PM) process. Water-atomized and vacuum-annealed powders of T15 HSS along with other ingredients like Zn-stearate (2%) and alumina (Al2O3) were used to produce the components. The percentage of alumina, sintering temperature and sintering time were considered as the controllable process parameters while the hardness of the sintered components was considered as the response variable. A statistical design of experiments known as central composite design (CCD) has been used to collect the data for developing the fuzzy model. The performance of the fuzzy predicting model has been compared with the response surface (RS) model with respect to mean squared error (MSE), mean percentage error (MPE) and co-efficient of determination (R2). It has been observed that the fuzzy model has much better predicting capability than the RS model.