Determination of temperature difference in squeeze casting hot work tool steel

The effects of process parameters on temperature difference (ΔT_RECS) of the squeeze casting part were studied by employing artificial neural network (ANN) and Procast software when cooling up to recrystallization temperature, such as interface heat transfer coefficient of metal/cavity die (h1), applied pressure (Ap), interface heat transfer coefficient of metal/male die (h2), die pre-heat temperature (Dt) and pouring temperature (Pt). An ANN model on the relationship between the processing parameters and ΔT_RECS was constructed. The test results on performance of the trained network show that the ANN model can predict ΔT_RECS with reasonable accuracy. Employing the ANN model the following conclusions could be made from this work: as far as the influence of process parameters on ΔT_RECS are concerned, the most important and the secondary parameter are Dt and Ap. H1 and Ap increase within a certain range, ΔT_RECS are found to increase. When Ap and h1 are above their respective critical point, the ΔT_RECS decreased rapidly. Pt, h2 and Dp increase within a certain range, ΔT_RECS are found to decrease.