|
How to improve the quality of strip steel plate manufacturing enterprises are facing a major issue. Hydraulic bending roll is one of the most common and most effective means of shape control , has been widely used in the mill flatness control. Therefore , a comprehensive in-depth study of shape control theory, precise hydraulic bending roll shape control system model , flatness control technology has become an inevitable requirement. This paper systematically studied the hydraulic bending roll control system theory and basic composition , the establishment of a hydraulic bending force control system mathematical model . Detailed analysis of the characteristics of the system , steady state error , the impact of interference on the system . Using Matlab Simulink software tools to build a hydraulic bending roll control system dynamic simulation model of hydraulic , bending force simulation, component parameters discussed the impact on the system characteristics . Currently, the mill hydraulic bending roll control system for the majority of the traditional PID control algorithm, however, the traditional PID control has proved difficult to meet the requirements of high-precision control . Firstly, genetic algorithm using adaptive on-line PID parameters are optimized , and then using the CMAC ( cerebellar network ) and PID compound control strategy to further improve the hydraulic circuit for hydraulic bending rolls , bending force tracking performance , and traditional PID method, system performance has been greatly improved, while greatly accelerate the speed of response of the system to improve the stability of the system , and achieved good control effect. In order to meet the requirements of high-quality products on the strip , the need to further improve the control precision shape , however , since the rolling process , the control object with nonlinear , multivariable , strong coupling characteristics, the traditional methods are difficult to establish a precise mathematical plate -shaped model . To solve the above problems , this paper ANFIS toolbox of Matlab software platform, Intelligent Modeling Approach ANFIS ( Adaptive Neuro-Fuzzy Inference System ) to achieve a bending force and the secondary plate -shaped fuzzy modeling . And the use of a large number of actual field data FIS rolling models were tested , the simulation results show that the method used to establish the ANFIS model is valid , the system is adaptive , anti-interference ability , shape error fluctuation is small, with high control accuracy.
|