Font Size: a A A

The Research Of Single Cold Rolling Mill Process Control System And Rolling Force Prediction

Posted on:2013-05-28Degree:MasterType:Thesis
Country:ChinaCandidate:M TongFull Text:PDF
GTID:2231330371997136Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
In recent years, with the continuous development of the metallurgical industry, cold-rolled strip steel products have gained more and more applications in the social applications, life and other aspects. Because of its flexibility and less investment, single stand has been widely used in the production of cold rolled strip steel. Meanwhile single stand process control system has a great influence on the quality of the produced strip steel, which is a key link in the entire cold rolling process and is an indispensable link. In this paper model setup and computing have been studied in single stand process control system, and a method combining neural networks with mathematical models is used for the research on the rolling force prediction. Besides, offline simulation has been applied. The main contents are as follows:This paper starts from the primary applications of the process control system, and uses a variety of load distribution models combination with strip steel rolling plan to determine the production of the pass and pass reduction. Based on the classic load distribution solution, load distribution of power balance model has been conducted by a method which takes advantage of its speed and pass reduction. Moreover according to reversing characteristics of single stand and layer table data of strip steel, a tension system can be determined to make rolling effective as much as possible. After completing setup, calculation of the mathematical model is conducted by the empirical formula and the ideal model. As rolling force is the main data of the mathematical model in the single stand control system, the rolling force prediction is studied. Then by using the exponential smoothing method for the rolling force prediction, adaptive learning is conducted, and the simulation comparison is also done. Finally through a method combining neutral networks with mathematical models, the rolling force is predicted. Besides the simulation comparison is done as same as above.Whether the setup calculation of a single stand mathematical model or the prediction of rolling force, they are both in good agreement with the actual strip steel data of a factory, and have high accuracy, which are suitable for on-line applications and guidance.
Keywords/Search Tags:Cold Mill Process Control, Neural Networks, Rolling Force Prediction
PDF Full Text Request
Related items