Research On Vibration Suppression Of F3stand Of 1580 Hot Tandem Rolling Mill

With the implementation of the Made in China 2025 strategic plan,the continuous upgrading and transformation of the manufacturing industry,and the rapid development of the steel industry,high quality,high efficiency and low pollution requirements are put forward for the production of plate and strip.The vibration of rolling mill will affect the production efficiency and product quality.During operation,the vibration will not only cause vibration marks on the surface of roll and rolling piece.It will also cause the fluctuations in the thickness of the strip,limit the rolling speed,reduce the production efficiency,and even cause safety accidents such as steel breakage and production shutdown.With the rapid development of rolling mill equipment towards automation and intelligence,the requirements for equipment vibration are more stringent.It is a hot research direction of rolling mill equipment vibration to conduct in-depth exploration of equipment vibration combining with machine learning technology.The phenomenon of rolling mill vibration is complicated,and there are many factors that affect rolling mill vibration.It has great limitation to adjust the equipment state according to experience to suppress equipment vibration.In the production process,the equipment operation status information is recorded in the PDA(Process Data Acquisition)system.Combined with the vibration mechanism model and the analysis and mining of the data,the vibration prediction of the rolling mill equipment in all directions is realized,and the vibration of the equipment is suppressed by optimizing the rolling schedule.Aiming at the serious vibration of 1580 hot strip mill,the vibration signal of the mill was collected by sensors.The mechanism models of vertical,horizontal and torsional vibration are established respectively by using mass and dynamic balance equations.The rolling process parameters and vibration data are analyzed to build the data prediction model.The vibration of rolling mill equipment is predicted and analyzed.The method of optimizing rolling process parameters is realized by using the data model to suppress the vibration of equipment.Theoretical model of rolling mill dynamics is established for horizontal and vertical vibration,the internal mechanism and transmission mechanism of rolling mill production are described,and the preliminary vibration control scheme is proposed.By using dynamic signal test and analysis to collect data and principal component analysis to process data in rolling process,the Bayesian Neural network prediction model is realized,and the integrated learning is used to improve the prediction accuracy and realize the accurate prediction of F3 frame vibration.Based on the data prediction models,combined with the vibration parameters in the mechanism model,the main process parameters are analyzed qualitatively and quantitatively.Verify the feasibility of rolling vibration suppression,and put forward the optimization scheme of rolling process from the aspects of mechanical design,equipment status,process parameters,etc.