Stacking ensemble transfer learning based thermal displacement prediction system

In the precision machining industry, machine tools are usually affected by various factors during machining, and errors generated accordingly. Where thermal error is one of most common difficult to control for tools. Therefore, in this study, six temperature sensors an eddy current displacement meter provided a tool with 4-axis dataset collection required model training, then data organized normalized. Next, introduced into variety learning models validated k-Fold cross-validation predicting those nonlinear that affect errors. At end, predicted results summarized compared find out best two better predictive performance pre-trained transfer learning. It observes from retraining conducted through applying Multilayer Perceptron (MLP) on these models, wherein MAE value as 0.40, RMSE 0.52625 R2 score 0.99696 respectively.