During the hot rolling process,the accurate prediction of finishing delivery temperature is crucial for controlling the three-dimensional dimensions and performance of the product.In order to improve the prediction accuracy of finishing delivery temperature,a data-mechanism fusion prediction modeling method was proposed.This method improves the data quality by fusing the results of the mechanism model into the data set based on hybrid feature selection,and introduces the Harris hawks optimization algorithm to optimize the prediction model,achieving high-precision prediction of finish-ing delivery temperature.The calculated results show that the optimized fusion model achieves EMA(mean absolute error,MAE),EMS(mean square error,MSE),and R2 of 4.136 8,31.97,and 0.932 2,respectively,and the proportion of data with prediction deviation within 15 ℃ is improved from 94.33％to 98.25％.Therefore,the proposed method can achieve high-precision prediction of the finishing delivery temperature in hot rolling.
hot continuous rollingfinishing delivery temperaturefusion modelintegrated modelingHarris hawks optimization(HHO)
NETL
NSTL
万方数据
维普
