Prediction and optimization of dynamic rolling force in strip rolling process driven by data-mechanism cooperation

A novel dynamic rolling force modified model driven by data-mechanism cooperation is proposed, which improves the prediction accuracy and rolling process stability. Firstly, the single-stand strip dynamic rolling force is predicted by traditional mechanism model and modified model respectively. The modified model has a greater probability of making the prediction error closer to zero, and the prediction root mean squared error of the modified model is lower than that of the mechanism model. Subsequently, targeting the strip dynamic rolling force with strength grade M3A30, the prediction effect of the modified model with strength grades M3A33 is better than that of the modified model with strength grades 1CD61. Finally, to reduce dynamic rolling force fluctuations during tandem rolling process, the rolling schedule is optimized before production. The maximum dynamic rolling force difference of S1, S2, S3, and S4 stands are reduced by 24.372 %, 29.071 %, 10.616 %, and 7.447 % respectively.