摘要
常规PID控制器是在冷轧机厚度控制中应用广泛的一种方法,但当AGC(厚度自动控制)系统特性或运行条件发生变化时,需要重新整定控制器参数,才能保证系统正常运行,使系统处于最佳工作状态。针对冷轧机AGC系统中的滞后环节,运用神经网络的自学习能力在线调整积分控制器参数值,该神经网络的权值与积分参数值相对应,可根据被控系统的动态特性调整积分参数,提高了调节器的自适应能力。
Conventional PID controller has been widely applied in the field of automatic gauge control （AGC） of the cold strip rolling mill. But when the characteristic of the AGC system or outside conditions change, the parameters of PID controller must be adjusted to guarantee the system works under good condition. In this thesis, the parameters are adjusted by the neural network according to the dynamic performance of the process in gauge con- trol system of the cold strip rolling mill. Because the weights of the neural network correspond with the PID parameters, the adaptive ability of the controller is improved.
出处
《冶金设备》
2009年第2期14-17,共4页
Metallurgical Equipment
