Font Size: a A A

Study On The Width Control Model And The Structure Dynamics Of 4200 Mill

Posted on:2011-07-19Degree:MasterType:Thesis
Country:ChinaCandidate:H L WuFull Text:PDF
GTID:2121330332458825Subject:Mechanical Manufacturing and Automation
Abstract/Summary:PDF Full Text Request
With the rapid development of metallurgical industry and the continue increase of automation technology, the quality and accuracy of the steel plate was required to improve by enterprises. However, the conventional order of medium plate mill can't meet the requirements of engineering practice, and backward control model has failed to achieve the prediction accuracy, which results in an urgent need to improve automation of the rolling process. As frequent equipment failure caused by vibration, therefore, it's necessary to develop new control methods and technological process of rolling and explore an effective way to reduce production failures and improve production efficiency.The width control model of 4200 mill was established on the base of RBF neural network combining the RBF neural network with the mathematical formula. Through the optimization of model parameters, the prediction accuracy of the network was improved and the overall error was controlled less than 4%. Compared with BP network, the high accuracy and stability of predictive control advantages of the RBF neural network were showed. In order to analyze the initial slab width and roll diameter on the impact of networking, six RBF network model were created on the different data packets conditions, which improved the Control model accuracy of rolling workshop.The phenomenon of main transmission of vertical rolling mill was introduced, non-linear model of vertical roller mill was established based on the analysis of the physical model and mechanical model, the fourth order Runge-Kutta method was used to solve the problem, The bifurcation diagram of the angular displacement changes with the angular frequency was simulated by the MATLAB software, the phase diagram and the poincare sections were obtained under different conditions of system, found that the system reached a periodic motion, quasi-periodic motion and other laws, it's provide an effective theoretical basis for normal operation of the device.The body modal analysis of the 4200 medium plate mill were explored finally, three-dimensional model was set up by Solidworks software and the finite element analysis software ANSYS10.0 was used. Frame and bearing roll of the 4200 medium plate mill were solved by subspace method applying the guidance of modal analysis theory, the rack's first 12-order vibration modes and roll's first 10-order modes were obtained. By analyzing the frequency of the greatest impact on the vibration deformation, a number of effective control measures was made in order to reduce equipment failure rates and increase productivity.
Keywords/Search Tags:Mealum Plate mill, RBF neural networks, COntrol model, NOnlinearvibration, Finite element, Modal analysis
PDF Full Text Request
Related items