Prediction of Peak Curing Temperature of Epoxy Resin Based on RBF Neural Network

In the manufacturing process of epoxy resin casting parts, such as dry casing, GIL, and GIS, due to the exothermic reaction in the two-stage curing process, it will produce temperature peaks when the curing degree changes sharply. If the temperature is too high at this time, it will have an impact on the internal stress of the composite material and affect the overall mechanical properties of the equipment. To improve the reliability of the equipment, it is necessary to predict the temperature peak in the 2-hold cycle. An RBF neural network was used to form a surrogate model of the temperature field, and the two temperature peak data sets from the 2-hold cycle were trained and tested for prediction. Using RBF neural network, it was able to have a good prediction of the peak temperature, and the RMSE of the prediction results are 0.0084 and 0.0442. It shows that the neural network can effectively reflect the peak temperature of the 2-hold cycle.