Artificial neural network prediction of the microstructure of 60Si2MnA rod based on its controlled rolling and cooling process parameters

An artificial neural network (ANN) model was developed for prediction of the microstructure of the 60Si2MnA spring steel rod based on its controlled rolling and cooling process parameters. The ANN predicted results show that the interlamellar pearlite spacing, the ferrite content and the average ferrite grain size all decreases rapidly with the increase of the cooling rate from the rod layer to the coil collecting station, and the effect of the rod-laying temperature and the finishing temperature on the microstructure is not obvious. The ANN model can predict the microstructure whether the process parameters interact or not, so it is critical for the quality control of the 60Si2MnA rod and will be widely used in its controlled rolling and cooling process.