磨浆机的专家控制  

Expert Control of Refine

作　　者：杨洋[1] 孟彦京[1] 

机构地区：[1]陕西科技大学,陕西省西安市710021

出　　处：《微计算机信息》2008年第25期307-308,共2页Control & Automation

基　　金：基金项目:"矩阵变换器馈电电机驱动系统直接转矩控制及DSP实现"(06JK350);颁发部门:陕西省教育厅

摘　　要：目前,磨浆机的传统控制方案为在纸浆浓度和流量稳定的基础上,通过检测磨浆机主电机功率传感器传送上来的功率信号,动态地调节磨浆机的刀距从而实现对纸浆的恒功率打浆控制。本文在此基础上提出了基于专家控制思想的磨浆机进退刀量控制方案。专家控制器具有一定的自学习功能,结合自学习算法自动修改不合适的输入输出量化值,以便更准确地控制磨浆机的磨刀进退位移量。此控制方案经过在纸厂的实际使用表明,在继承了传统控制算法平稳精确的打浆控制基础上,大大提高整个系统的响应速度和鲁棒性能。At present, the traditional control of refin er is that on the bases of stable paper pulp density and flux, through examing the power signal of the main motor power sensor of refiner, adjust the clearance of refiner dynamically, so as to realize permanent power control of paper pulp. This article proposed refiner knife onset and retreat control based on expert control. The expert controller has certain self-learning function, combined with self-learning algorithm, revises inappropriate input and output quantification value automatically, in order to accurately control refiner knife’s onset and retreat displacement. The actual use of this method indi-cated that it not only inherited steady and precise algorithm of traditional paper pulp control, but also greatly enhanced response speed and robustness of the overall system.

关 键 词：专家控制 磨浆机 自学习 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]