Font Size: a A A

The Research Of Temperature Process Control Model In Finish Rolling

Posted on:2006-12-03Degree:MasterType:Thesis
Country:ChinaCandidate:G F XiaoFull Text:PDF
GTID:2121360182461339Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
2050 roll line in some large steel factory has been put into production for over ten years, facing with great reconstruct. Especially, there was a long way between our strip temperature predict level and the international first-class level. The process control model of hot rolling is the kernel technology in hot rolling. We intended to improve strip temperature precision by developing the forecasting effect of process control model in finish rolling.Due to two reasons, the finish rolling precision process control was a very complex job. Firstly, it not only refer to a lot of physical parameters, such as water-quantity, run-out speed, strip-thick, but also refer to yield deformation, the interior structure and performance variation of strip. Secondly, the finish rolling process control system is a nonlinear, strong coupling and indetermination system, so traditional modeling technology could hardly improve the precision of hot rolling model any more.Modeling method and 2050 temperature system is briefly introduced firstly in the paper, mechanism model of air cooling, water cooling, heat transition in rollers and rolling deform heat on temperature drop in finish rolling is built according to fact in this steel factory, uncertain parameters are nonlinear regression. The simulation system built in MATLAB realizes forecasting temperature of finish rolling point by point and improves its forecasting precision.Because traditional modeling technology could hardly improve the hot rolling model's precision any more, a compositive temperature forecasting model combined mechanism model with BP neutral network is put forward in the paper. Offline simulation effect shows that forecasting precision has been improved dramatically, proving that neutral network is feasible to modeling on nonlinear complicated system. Meanwhile, an intelligent optimization-seeking algorithm, PSO, has been stated, and this algorithm improved the performance of neutral network by applying it to neutral network. Some referential thoughts and ideas have been put forward for BP neutral network, laying a favorable base for furthering application online.
Keywords/Search Tags:hot rolling, temperature control, PSO algorithm, BP network, intelligence technology, modeling
PDF Full Text Request
Related items