An artificial neural network (ANN) optimized by genetic algorithm (GA) is an established prediction model of bending force in hot strip rolling. The data are collected from factory of steel ...manufacture. Entrance temperature and thickness, exit thickness, strip width, rolling force, rolling speed, roll shifting, target profile, and yield strength of strip are selected to be independent variables as network inputs. MATLAB software is utilized for establishing GA-ANN model and achieving the purpose of obtaining the bending force as results of setup model, as well as the GA method is used to optimize the initial weights and biases of the backpropagation neural network. Mean absolute error (MAE), mean absolute percentage error (MAPE), root mean squared error (RMSE), and correlation coefficient are adapted to evaluate the performance of the model. The predictive results are compared with the measured results to verify the accuracy of the GA-ANN prediction model. It is found that the optimization effect is the best with the population size 40 crossover probability of 0.7 and the mutation probability of 0.05 at the same time, the fitness function value can reach 80.7. In addition, the ANN architecture 9-11-1 trained with Bayesian regulation “trainbr” function has the best performance with mean absolute error of 0.01 and correlation coefficient of 0.983. With a deeper understanding of neural networks through the analysis of the GA-ANN model, the proposed model can be flexibly used for on-line controlling and rolling schedule optimizing.
Strip crown prediction model based on support vector machines (SVMs) is proposed, and the improved adaptive mutation particle swarm optimization (AMPSO) algorithm is used to optimize the model ...parameters (C,σ) in this paper. A set of online inspection data from the steel plant is adopted to train and test model. The overall performance of the model is evaluated by the decision coefficient (R2), mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE). Comparing the results calculated by other relevant models and the present model with same experimental data, the accuracy of the model presented in this paper is verified. The AMPSO‐SVR model can be successfully applied to the prediction of strip crown in hot rolling. Both prediction accuracy and generalization ability of the new model have achieved good results. The proposed model provides a new method and idea for shape control and optimization research in hot strip rolling process.
A new predictive model for strip crown in hot rolling by using the hybrid AMPSO‐SVR‐based approach is presented. It is a new model combined with industrial big data and artificial intelligence. Both prediction accuracy and generalization ability of the new model have achieved good results. The proposed model provides a new method and idea for shape control and optimization research in hot strip rolling process.
As the basis for a flatness control system, flatness actuator efficiency describes an actuator’s control ability, but it is difficult to obtain an actuator efficiency factor accurately through ...rolling tests because of the complicated subsidiary facilities of the mill. This paper proposes a novel simulation approach that is applied to obtain the actuator efficiency factors in terms of work roll bending, intermediate roll bending, and intermediate roll shifting for a six-high Universal Crown Control mill (UCM mill). A three-dimensional (3D) finite element model of the mill was developed to simulate the dynamic strip rolling process. The validation of rolling experiments shows that this model has enough precision, where the relative error of strip thickness between simulated values and actual values is less than 1.0%. The effects of the actuators on strip thickness profile, crown, edge drop, and elongation difference of longitudinal fibers were investigated. In the case of different actuator parameters, the curves of actuator efficiency factors were obtained and quantitatively descripted by truncated Legendre orthogonal polynomials. The mechanism of flatness control was studied based on an analysis of the actuator’s influence rule on the elastic deflection of rolls and 3D distribution of rolling pressure. The results indicate that the curves of actuator efficiency factors have a symmetrical upside-down v-shaped distribution and contain the quadratic and quartic flatness components. The actuator efficiency factors of intermediate roll shifting have a nonignorable variation with the change of actuator parameters. This study is the first attempt to obtain actuator efficiency factors for UCM mill using an elastic–plastic finite element method.
The tandem cold-rolling process is a multivariable, nonlinear, and strongly coupled complex control procedure, in which the key technologies of automatic gauge control (AGC) and automatic tension ...control (ATC) are extremely comprehensive, and high precision is required. This paper analyzes the rolling characteristics of tandem cold-rolling process and proposes an innovative multivariable optimization strategy based on inverse linear quadratic (ILQ) optimal control theory for thickness and tension control. First, a new state space model of the tandem cold-rolling process was introduced and verified based on the basic equations of rolling technology and field data. Then, meaningful influence rules on the complex rolling process were obtained by analyzing rolling characteristics. For the complex rolling process, a novel ILQ control strategy was introduced into the thickness and tension control system. As a result, by a series of experiments, the effect of disturbance on the thickness and tension was attenuated to an arbitrary degree of accuracy through the proposed control strategy. Simulation results showed the excellent control performance of the proposed ILQ control strategy compared with the conventional proportion and integration (PI) control strategy.
To improve the smart manufacturing capabilities of strip hot rolling, based on digital twin (DT) and cyber-physical system (CPS), this paper proposes a data-driven approach for diagnosing hot-rolled ...strip crown. Since the hot rolling process features heredity, nonlinearity and strong coupling, the diagnosis of strip crown is an imbalanced problem with ill-defined decision boundaries. Conventional regression methods tend to learn more information from the majority class, which ignore the strip with unqualified crown. To address this challenge, a hybrid multi-stage ensemble model (HMSEN) is presented to classify strip crown. Initially, a novel data-resampling method that combines adaptive synthetic sampling (ADASYN) with repeated edited nearest neighbor (RENN) is proposed to assign more attention to unqualified crown. Subsequently, using the reinforced data, a multi-stage ensemble model is built to enhance the classification performance. Furthermore, the best-performing HMSEN is identified by exploring various combinations of base classifiers. The experimental results demonstrated the proposed novel resampling method outperforms comparison methods on crown dataset. Significantly, the proposed HMSEN outperforms not only the existing regression models but also the mechanism model. Therefore, HMSEN is the most robust and effective model to intelligently diagnose hot-rolled strip crown with unbalanced data.
Flatness plays a crucial role in determining the quality of products in strip cold rolling. Data driven methods have shown promise in flatness prediction by effectively capturing the nonlinearities ...and strong coupling present in cold rolling processing, surpassing the capability of conventional methods. However, existing data driven models remain restricted by a lack of rolling theory guidance, a black-box nature of predictive processes, and gradient conflict of multi-channel flatness. To overcome these limitations, this paper proposes an interpretable mechanism guided multi-channel distributed meta learning framework for flatness prediction. Initially, significant physic-based parameters, such as theoretical rolling force deviation and tension deviation, and controller parameters are calculated to guide data driven modeling. Subsequently, a distributed meta learning framework is modeled for multi-channel flatness to eliminate gradient conflict. Furthermore, eXplainable Artificial Intelligence (XAI) technique is implemented to ensure the transparent predictive processes of multi-channel flatness. The analysis results present that theoretical parameters and controller parameters effectively improve the performance of flatness prediction. In addition, the comparative results demonstrate that the proposed framework outperforms the existing flatness prediction methods and other state-of-the-art machine learning methods by 4.24%. Importantly, the XAI-based explanation of the proposed framework effectively enhances the credibility of data driven flatness prediction.
Display omitted
•An interpretable distributed meta learning framework is proposed for flatness prediction.•The interpretability and performance of data driven model is enhanced by integrating rolling mechanism.•Distributed learning framework is effective in eliminating gradient conflict of multi-channel flatness in training procedure.•The predictive process of data-driven model in cold rolling is first explained by eXplainable Artificial Intelligence (XAI).
The prediction precision of mechanics parameters such as rolling force and torque affects the yield, quality, cost, and benefit of products during the cold strip rolling. The lever arm coefficient is ...the core linking rolling force and torque, but there is no mathematical model of this in cold rolling. The distribution of rolling pressure and the change rule of lever arm coefficient under different reduction, forward and backward tension stress, deformation resistance, and friction coefficient in cold rolling are illustrated based on 3D (three-dimensional) elastic-plastic FEM (finite element model) simulation. The mathematical model of lever arm coefficient is built according to online measured data processed by BP (back propagation) neural network in the tandem cold rolling plant. The predicted rolling forces calculated on the basis of this model and upper bound method are consistent with online measured values. The proposed model provides valuable guidelines to determine the reduction and check the strength of the equipment such as rolls and stands.
In hot strip rolling process, rolling schedule setup, geometrical accuracy (thickness and profile), and even the final product homogeneity of mechanical properties are affected by the automatic ...control, and the rolling force and torque are the prerequisite in the control process. A new cosine velocity field is firstly proposed in this paper to get the values of the required minimum rolling force and torque. The field and equal area (EA) yield criterion are used to integrate the internal plastic deformation power. Using the co-line vector inner product method, the friction power is analyzed. Finally, the analytical expressions of rolling force, rolling torque, and stress effective factor are obtained. The theoretical predictions of rolling forces are compared with on-line measured ones in a hot strip rolling plant and other researchers’ models. Results show that the calculated rolling forces are in fair agreement with the actual measured ones, and the proposed solution is considered to be applicable for solving hot strip finish rolling.
The present investigation devises a flatness control strategy for a five‐stand tandem cold rolling mill. In addition, intermediate roll shifting (IRS)–induced rigidity characteristics of the six‐high ...Universal Crown Control mill (UCM mill) as a key model for the automatic thickness and flatness control systems are established. A three‐dimensional elastic‐plastic finite element analysis for a strip rolling process is conducted when the UCM mill is subjected to different IRS values. The convergence and precision of the simulation model are verified using experimental data in a five‐stand tandem cold rolling line. Through a steady‐state rolling analysis of the process, the vertical and transverse rigidity characteristic curves of the roll‐stack, reduction strain field of the strip, variation in strip crown, elastic deflection of the rolls, and contact stress field between rolls are extracted. The results show that the IRS obviously changes the deflection curves of the rolls and then causes the variations in the exit thickness and transverse thickness difference of the strip. The nonuniform stress distribution between rolls caused by the increasingly IRS value lead to local stress concentration on the side of the roll. A mechanism for the effect of IRS on the rigidity characteristics of the mill is presented and discussed.
The dependence of rigidity characteristics of UCM mill on intermediate roll shifting (IRS) are investigated using three‐dimensional elastic‐plastic FE analysis, as shown in the Figure. The effects of IRS value on rolling force, strip reduction strain field, roll elastic deflection, and contact pressure between rolls are revealed in detail. Further, the vertical and transverse rigidity characteristic curves of the roll‐stack affected by IRS are obtained and discussed.
The computational complexity of privacy information retrieval protocols is often linearly related to database size. When the database size is large, the efficiency of privacy information retrieval ...protocols is relatively low. This paper designs an effective privacy information retrieval model based on hybrid fully homomorphic encryption. The assignment method is cleverly used to replace a large number of homomorphic encryption operations. At the same time, the multiplicative homomorphic encryption scheme is first used to deal with the large-scale serialization in the search, and then the fully homomorphic encryption scheme is used to deal with the remaining simple operations. The depth of operations supported by the fully homomorphic scheme no longer depends on the size of the database, but only needs to support the single homomorphic encryption scheme to decrypt the circuit depth. Based on this hybrid homomorphic encryption retrieval model, the efficiency of homomorphic privacy information retrieval model can be greatly improved.