Study On Shape Control For Cold Rolling Of Effective Matrix Based On Weighted Intelligent Method

Along with the development of national economy and the improvement of living standards, the demand of strips is growing unceasingly, simultaneously the requirements of the product’s quality of strips is also increasing. Flatness is one of the most important quality indexes of strip steel, and flatness controlling technique is the hot topic in the rolling area. In recent years, artificial intelligence has been widely used in industrial process for its advantages in modeling, optimization and control.Firstly,taking 1050 six high rolling mill for example and using tilting roll, bending work roll, asymmetry bending work roll and middle roll shifting intermediate roll as flatness control means, the effective coefficients under conditions of different strips of the linear flatness, the quadratic flatness, the cubic flatness and the biquadratic flatness for different control means are calculated by theory model. Then the effective rules of effective matrix for different flatness control means are showed, which provides the foundation for the building and realizing on-line flatness control.Secondly, for the problem of the traditional neural network training time is long and easy to fall into local minimum, extreme learning machine(ELM)neural network is proposed to predict strip flatness. Considering the random input weights and bias values of neural network may influence prediction accuracy, the differential evolution(DE) algorithm is used to optimize ELM neural network.Because the traditional effective function method and static flatness effective matrix is not good enough, dynamic effective matrix method for flatness control by analysis and operation of the massive production data is proposed and the flatness forecast model is established. The on-line dynamic effective matrix is calculated to realize the effective control of flatness.Finally, the simulation application is carried on the 1050 six high rolling mill using the dynamic effective matrix method for flatness control. The results confirm fully the validity of the flatness forecast model and the dynamic effective matrix method for flatness control.