Font Size: a A A

Intelligent Control System Of Tire Tread Two Compound Extrusion Line

Posted on:2022-07-08Degree:MasterType:Thesis
Country:ChinaCandidate:Y H XuFull Text:PDF
GTID:2492306335466604Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
The tire industry is an important basic industry in China.In order to ensure the safety of automobile’s driving,the tire industry puts forward very strict requirements on the quality of tire products.Tread is an important part of the tire,so its product quality is particularly important.The quality of tread products is mainly related to three parameters:weight,shoulder width and total width.By controlling the weight,shoulder width and total width of the tread,improving the product quality of the tread is of great significance to improve the tire’s quality and ensure the safety of the car.In order to solve the problem of controlling the tread’s weight,shoulder width and total width intelligently,this paper mainly does the following work:(1)The key technology of on-line accurate intelligent perception of shoulder width and total width of tread is proposed and realized.The tread’s surface is dark and in motion on the production line.The common machine vision technology is limited by lighting and other conditions,so it is difficult to satisfy the application scenario in this paper.In this paper,three-dimensional line laser profile sensor is used to detect the tread size in motion on line,which has high detection precision and few environmental impact factors.By finding the curvature extremum on the contour,the feature points can be found and the tread size information can be obtained.In view of all kinds of abnormal conditions that may be encountered in the actual tread production process,according to certain rules,the data and calculation results are filtered,the measurement of tread size is stable,and the detection precision reaches 1 mm.(2)The method to determine the standard values of controlled variables of tread is proposed and realized.The controlled variables of tread refer to the shoulder width and total width of the tread at the front of two compound extrusion line,as well as the tread’s meter weight of the continuous scale.To realize the intelligent control of tread’s weight,shoulder width and total width,it is necessary to compare the corresponding standard value with the actual value.This paper collects the process data of tread production and uses the regression algorithm in machine learning to establish a mathematical model According to the process standard values of the tread cut finally,the standard values of the tread’s weight,shoulder width and total width at the front of the production line are determined to provide the set points for the control system.(3)To solve the control problem of complex plant with the characteristics of three-input and three-output,strong coupling,strong time-varying and nonlinear elastic deformation of rubber material of tire tread two compound extrusion line,a model-free control method of compact-format dynamic linearization.(CFDL)for MIMO with parameter self-tuning for pseudo Jacobian matrix is proposed.Considering the characteristics of the two compound extrusion line above,it is difficult to establish an accurate dynamic mathematical model by mechanism modeling or system identification.Therefore,this paper proposes a model-free controller based on data-driven.At the same time,aiming at the parameter tuning problem of model-free controller,an online self-tuning method of pseudo Jacobian matrix parameters based on BP neural network is proposed to solve the problem that the selection of model-free controller’s parameters depends on artificial experience.(4)The intelligent control system of tread two compound extrusion line has been applied and verified in the industrial scene.The intelligent control system of tire tread two compound extrusion line is developed by using the key technologies of intelligent perception of tread size,determination method of standard values of tread’s controlled variables,CFDL based on model-free control method for MIMO with parameter self-tuning for pseudo Jacobian matrix.The three inputs of the system are two extruders’speeds and the production line’s speed,and the three outputs are tread’s weight,shoulder width and total width.The intelligent extrusion system has been successfully tested in the production site.The tread produced meets the process requirements,and the product quality has been significantly improved.
Keywords/Search Tags:Tire tread, Two compound extrusion line, Intelligent perception, Model-free control, Multi-input and multi-output system
PDF Full Text Request
Related items