基于神经网络的解耦控制在板带钢生产过程中的应用

The Application of Decoupling Control Based on Neural Network During Production of Plate Belted Steel

根据板带钢轧制工艺的特点和板形板厚的控制要求,本文提出了一种基于神经网络的解耦控制方法,并设计了具有自适应、自学习解耦能力的智能解耦控制器。实际运用表明,该方法能够实现板形板厚的解耦,满足板带质量控制要求并具有良好的鲁棒性。

According to characteristic of the rolling technology of plate belted steel and the control requirement to steel shape and plate thickness, this paper proposes a decoupling control method based On neural network, and designs an intelligent decoupling controller with self-adapton and self-study decoupling ability. Practice proves this method can implement decoupling of steel shape and plate thickness, meets the requirements of steel plate quality, with excellent robustness.