Search

Machine learning performs well in many problems. However, the tendency to generate predictions that violate theoretical knowledge makes it difficult to apply to practical processing. To resolve this situation, this paper combines domain knowledge with a data-driven model, proposes a theory-guided machine learning framework based on a parameter transfer strategy, and applies it to the width prediction of plates after multiple passes of hot rolling. The framework applies a swarm optimization algorithm to the original theoretical model and generates numerous highly-physical consistent samples. The established deep neural network (DNN) model is trained with simulated data, and the parameters are fine-tuned using a parameter transfer strategy combined with actual data to ensure excellent adaptation to the actual environment based on adequate learning of theoretical knowledge. In tests, the proposed model had the best overall prediction performance in this paper. Meanwhile, the developed model is consistent with the existing perception of rolling theory. This allows for the quick and reliable application of machine learning models in production. [Display omitted] • A novel theory-guided machine learning framework for width prediction of rolled plates. • Pre-training of the model using the data generated by the optimized theoretical model. • The proposed model has the best accuracy in both sufficient and less actual samples. • The variable influence of the proposed model is consistent with theoretical perceptions. [ABSTRACT FROM AUTHOR]

During the slab sizing press (SP) process, the pressing force corresponds to the slab profile, which guides the production schedule design and the final profile control. To accomplish the prediction of pressing force for SP, an improved ensemble method based on chaotic Harris hawks optimizer (CHHO) and stacking is proposed. A mechanistic knowledge is introduced for feature selection that enhances the rationality of input features. Subsequently, 11 machine learning models are compared and 5 of them are selected as candidate learners for the stacking method. Based on the candidates learners, 8 stacking strategies are constructed, which the stacked model with extratree regressor, gradient boosted decision trees, and kernel ridge regression (KRR) as base‐learners and KRR as meta‐learner performs the best. The R2, MAE, mean square error, mean squared log error, and mean absolute percentage error for the test dataset are 0.9912, 0.0856, 0.0167, 0.0005, and 2.00%, respectively, and 95% of the prediction errors are less than 0.15 MN. Then, sensitivity analysis and predictive analysis based on Shapley Additive Explanations are performed to demonstrate the good alignment of the proposed model with physical reality. Furthermore, to cope with the complexity and uncertainty of the production process, the proposed CHHO‐stacking model and the kernel density estimation method are integrated to model the prediction interval. [ABSTRACT FROM AUTHOR]

The accurate prediction of strip crown is the precondition of the shape preset model in hot strip rolling. In this study, a new hybrid strip crown forecasting model is proposed in combination with extreme learning machine (ELM) and the industrial data. The production data of 1780 mm hot strip rolling are collected by the on-site data acquisition system to form dataset. Principal component analysis (PCA) is used to reduce the dimension of the input data for modeling samples. To improve the prediction accuracy of ELM, an improved PSO based on S-curve decreasing inertia weight (SDWPSO) is proposed to optimize the initial weights and biases of ELM. Finally, the optimal ELM model and simple production dataset are used to establish a strip crown prediction model of hot strip rolling named PCA-SDWPSO-ELM. The comprehensive performance of the proposed hybrid PCA-SDWPSO-ELM prediction model is evaluated by MAE, MAPE and RMSE. The superiority of the proposed model is also proved by comparing the prediction results with those of the other three comparison models. The research shows that the hybrid PCA-SDWPSO-ELM method can solve the problem of nonlinear and strong coupling in traditional engineering. It is suitable for parameter prediction and optimization in the iron and steel manufacturing industry, especially in the process of shape control in hot strip rolling. [ABSTRACT FROM AUTHOR]

In the hot strip process, the prediction of strip shape is a key factor to improve the quality of products. However, strip‐shape prediction models based on traditional machine learning algorithms with relatively simple network structures often rely on data preprocessing and feature selection. Herein, five strip‐shape prediction models are proposed by combining sparrow search algorithm (SSA), particle swarm optimization algorithm (PSO), extreme learning machine (ELM), and deep multilayer extreme learning machine (DELM). The prediction experiments show that the SSA‐DELM model has highest accuracy, and the determination coefficient (R2) of crown prediction and flatness prediction of SSA‐DELM model is 0.971 and 0.979. 100% of the samples have a prediction error of less than ±7 μm and ±7 IU. Furthermore, to solve the problem of concept drift affecting prediction accuracy in industrial processes, online optimization method is proposed. Especially, online sequential deep ELM model (OS‐DELM) optimized by SSA and guided by double window drift detection is proposed to fill the gap. Compared with other online optimization methods, the update time with this method can be reduced by up to 83.05% averagely, which is more suitable for online strip‐shape prediction in hot rolling process. [ABSTRACT FROM AUTHOR]

The flatness control system of cold-rolled strip is characterized by nonlinearity, strong coupling, and multivariable features. Theoretical models of flatness derived from physical mechanisms of the rolling process are intricate. This complexity makes it challenging to establish an accurate control model. Reinforcement learning (RL) is a machine learning method that uses an intelligent agent to learn from data interacting with its environment. In this study, considering the large quantity of historical data generated by the strip steel control system, we used the RL method to analyze the data offline in order to train a data-driven controller. We combined it with ensemble learning to propose a novel method for flatness control based on ensemble deep reinforcement learning (EDRL). Initially, we constructed a state space model for the cold-rolled flatness utilizing historical production data. This model served as the environment for offline RL training. We present an EDRL-based flatness control method that incorporates the principles of ensemble learning. Ensemble learning amalgamates multiple individual models to achieve superior performances. In the online simulation of rolling production, the EDRL-controlled flatness value was less than 5 (I-Unit) IU from the target flatness value, exceeding the performance of the proportional–integral (PI) controller. The proposed method surpassed the conventional approach in terms of both robustness and its capability to regulate flatness. • An EDRL control strategy is proposed for the cold rolling flatness control problem. • The integration strategy effectively improves the convergence speed of flatness. • The proposed control strategy outperforms state-of-the-art PI controllers. • The proposed integrated strategy exhibits robust stability in industrial scenarios. [ABSTRACT FROM AUTHOR]