综合智能控制策略在真空退火炉中的应用  

Application of Integrated Intelligent Control Strategy in a Vacuum Annealing Furnace

作　　者：于波[1] 强明辉[1] 余亚冰[1] 李晓斌[1] 

机构地区：[1]兰州理工大学电气工程与信息工程学院,兰州730050

出　　处：《电气自动化》2010年第3期25-27,共3页Electrical Automation

摘　　要：真空退火炉退火温度的精确控制是一个典型的大时滞、大惯性、存在强交叉耦合的、时变的复杂控制问题,采用常规的PID控制器很难满足对退火温度的精确控制。通过对退火炉工作机理及退火温度控制系统的分析,提出了一种综合智能控制方案;首先利用人工神经网络对真空退火炉进行建模,之后利用遗传算法对神经网络的权值和阈值进行优化;最后,将此模型作为一个预估模型投入到控制系统中去。在控制器的设计过程中,不再使用单一的控制策略,而是在生产过程中,根据退火工件温度的输出误差大小来随时改变控制器的结构,从而达到对真空退火炉退火状态的最佳控制。Precise control of annealing temperature for vacuum annealing furnace is a typical complex control problem for its large delay, large inertia, strong cross-coupling and time-varying, hence the conventional PID controller is incapable of realizing it. In this paper, an integrated intelligent control program is put forward with the analysis of the annealing temperature control system and the working mechanism of the annealing furnace： First of all, artificial neural network is applied to model the vacuum annealing furnace; Then, optimize weights and threshold value of neural network with the aid of genetic algorithm; finally, this model is used as a prediction model to go into the control system. In the designing process of controller, single control strategy is no longer applied, instead, to achieve the best control of the vacuum annealing furnace, the structure of the controller is modified at any time in the production according to the temperature output error of the annealing workpiece.

关 键 词：真空退火炉 智能控制 算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]