基于神经网络的板形板厚解耦控制的研究与应用

Design and Application of Strip Shape and Thickness Decoupling Controller Based on Neural Network

板带钢生产过程中,厚度和板形是两个重要质量指标。指标的高低直接关系到板带钢的质量。在实际控制中,由于板形和板厚耦合及控制的非线性和影响因素多,传统的解耦控制方法无法满足质量控制要求。本文提出了一种基于神经网络的解耦控制方法。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。

In this paper,intelligent strip shape and thickness decoupling control based on neural network is put forward and designed according to the craftwork characteristic of strip rolling and the decoupling of Strip shape and thickness.The real applications show that this kind of new controller can be good for implementing decoupling of strip shape and thickness and improving the quality of strip rolling,and good robustness.