The strip crown in hot rolling has the characteristics of multivariablity, strong coupling and, nonlinearity. It is difficult to describe accurately using a traditional mechanism model. In this paper, based on the industrial data of a hot continuous rolling field, the modeling dataset of a strip steel prediction model is constructed through the collection and collation of the on-site data. According to the classical strip crown control theory, the important process parameters that affect the strip crown are determined as input variables for the data-driven model. Some new intelligent strip crown prediction models integrating the shape control mechanism model, artificial intelligence algorithm, and production data are constructed using four machine learning algorithms, including XGBoost, Random Forest (RF), Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP). The overall performance of the models is evaluated using error indicators, such as Mean Absolute Percentage Error (MAPE), Root Mean square Error (RMSE), and Mean Absolute Error (MAE). The research results showed that, for the test set, the determination coefficient (R²) of the predicted value of the strip crown model based on the XGBoost algorithm reached 0.971, and the three error indexes are at the lowest level, meaning that the overall model has the optimal generalization performance, which can realize the accurate prediction of the outlet strip crown in the hot rolling process. The research results can promote the application of industrial data and machine learning modeling to the actual strip shape control process of hot rolling, and also have important practical value for the intelligent preparation of the whole process of steel.

Stainless steel/copper composite strips (SSCC) were fabricated using stacked rolling followed by finishing processes. These SSCC underwent annealing at temperatures of 300℃, 400℃, 500℃, and 600℃ for a duration of one hour. The deformation characteristics of the composite strips during rolling were thoroughly analyzed. In addition, investigations were carried out on their mechanical properties and microstructural evolution. The results indicate that as the annealing temperature increases, the tensile strength of the SSCC decreases slightly, while the elongation increases significantly. Initially, the peel strength of the rolled SSCC is the lowest, but it increases progressively as the annealing temperature rises. At the annealing temperature of 600°C, the peel strength of the composite strip reaches its maximum value of 7.1 N/mm. Significant interface delamination is observed during the bending process of the rolled SSCC. However, at the annealing temperature of 600°C, interface delamination is absent in the SSCC, resulting in optimal bonding quality. The rolled SSCC exhibits prominent defects, including cracks and holes at the bonding interface. In contrast, annealing significantly reduces both the number and size of these cracks and holes, indicating improved interface bonding quality. As the annealing temperature increases from 300°C to 600°C, the grain size of copper increases from 8.45μm to 28.5μm, and the proportion of low-angle grain boundaries (LAGBs) rises from 1.2% to 3.7%. Additionally, a significant number of 60° grain boundaries is observed, suggesting the formation of numerous annealing twins. In the stainless steel (SS) layer, martensitic transformation occurs due to rolling. The austenite grains in the SS progressively enlarge, while the martensite grains decrease accordingly as the annealing temperature increases.

For the analytical model of rolling force of ultra-thin strip, the iterative conditions of the exit plastic zone are improved to solve the convergence problem of the Fleck model in small reduction rolling. The nonlinear law of friction coefficient in multi-pass rolling is analyzed, and the friction coefficient database for sample data is established through the friction coefficient calculation model, which is used GWO-KELM neural network training friction coefficient prediction model, the Fleck rolling force prediction model based on the modified friction coefficient is established ultimately. A comparative analysis of prediction errors is conducted on three different specifications of strip steel using actual production data from a multifunctional 280 mm 20-high mill. The results show that the best performing MSE, RMSE, MAE, MAPE and R2, with values of 170.48, 13.06 kN, 9.01 kN, 3.30%, and 0.989, respectively. The accuracy of the modified rolling force prediction model is significantly improved, and the data scale of friction coefficient database can be continuously expanded, so the accuracy of the rolling force prediction model can be continuously improved.