Forming Process Development And Research On Mathematical Model Of Four-roll Bending-plate Machine

The four-roll bending-plate machine is an important large-scale mechanical equipment,which is widely used in many engineering industries.Its function is to roll metal sheet into cylindrical parts of various specifications to meet industrial needs.In this paper,W12 CNC four-roll bending-plate machine is taken as the research object,with the aim of improving the rolling effect.On the basis of the four-roll bending-plate machine round forming process,the online adaptive optimization algorithm of the rolling process is studied to improve the rolling accuracy and rolling efficiency;establish an elliptical cylinder rolling model,study the elliptical coil forming process,develop an elliptical coiling method suitable for industrial applications;establish a BP neural network roll gap prediction model for the roll gap value,and improve the four-roller rolling machine Intelligent level.First,when analyzing the round forming process of the four-roll bending-plate machine,based on the structural parameters of the W12four-roller rolling machine,the main steps of preparation,centering,pre-bending,rolling,closing and rounding in the rolling process conduct stress analysis and establish a theoretical model of the rolling process.Used Siemens PLC s7-300 as the controller to write the rolling circle model program,compared the calculated roll gap value with the actual roll gap value and analyze the accuracy of the model.When improving the accuracy of the coil model,the adaptive optimization algorithm is used to approximate the target value(roll gap),and the cylinder gap of the same specification of the same plate is dichotomized to approximate the roll gap value,and the different specifications of the same plate Lagrange interpolation method is used to predict the roll gap value of the cylindrical parts.When researching on the ellipse rolling model,due to the difference in the radius of curvature of each point of the ellipse,it is not conducive to rolling in the actual production process,nor to the compilation of the program.Therefore,the calculation process of the ellipse curvature radius is simplified,and the arclength of each arc of the ellipse is calculated by using the Simpson formula and so on,so as to realize the automatic rolling of the ellipse.The adaptive learning optimization algorithm is also introduced to optimize the ellipse model and improve the model’s calculation accuracy.A roll gap prediction control model for BP neural network is proposed for the rolling process.The data generated during the production process of the four-roller plate rolling machine is used as a training sample and trained through the BP neural network.After analyzing a large amount of actual data,a more accurate Neural network prediction model to replace the stress analysis of the circle model.This paper analyzes some shortcomings in the traditional coiling process,points out the possible reasons and proposes corresponding improvements,which provides some effective ideas for the development of the four-roll bending-plate machine.