Numerical optimisation of superplastic deformation

Based on an approach due to Padmanabhan and Davies, a multi-dimensional regression analysis has been developed which predicts the superplastic deformation parameters of m (the strain-rate sensitivity index) and K (the strength parameter) as functions of strain rate, grain size and temperature. Further analysis enables the optimisation of the operating conditions (for minimum power consumption) through a prediction of the external load and power consumption using the predicted values of m and K. The procedure has been validated by applying it for the analysis of the experimental data pertaining to the tin-lead eutectic alloy. It has been pointed out that the technique could be useful for problems (not necessarily in the area of superplasticity) where a particular parameter depends on a number of independent variables. © 1990 Chapman and Hall Ltd.