论文部分内容阅读
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。
Strip steel production process, thickness and shape are two important quality indicators. The level of indicators is directly related to the quality of strip steel. In the actual control, the traditional decoupling control method can not meet the quality control requirements because of the nonlinearity and influence factors of the coupling and control of plate shape and plate thickness. This paper presents a decoupling control method based on neural network. An intelligent decoupling controller with adaptive and self-learning decoupling capability is designed. The practical application shows that the method can realize the decoupling of the plate thickness, meet the strip quality control requirements and have good robustness.