基于神经网络内模控制的毛细管流变仪温控系统设计

Design of Neural Network Based Internal Model Temperature Control System for Capillary Flow Test Equipment

为了提高毛细管流变仪加热腔温度控制效果,在研究以多变量、交叉耦合为特征的控制对象的基础上,给出了一种神经网络内模控制算法。以通过学习得到的动态神经网络作为内部模型,将直接逆系统学习法得到的神经网络作为控制器,构造了MIMO控制系统。介绍了毛细管流变仪温度控制系统软、硬件的开发。实际运行结果表明,该系统控制精度高,鲁棒性好,可靠性高。

In order to improve the performance of the temperature control system for cube of capillary flow test equipment（TCSCCFFE）, a neural network internal model control algorithm was put forward, which is based on the research of the control object characterized with multivariable and cross coupling. The Multi-Input Multi-Output （MIMO） control system was constructed, as the internal model is dynamic neural network obtained from study, while the controller is neural network obtained from direct inverse-system study. The development of software and hardware system of TCSCCFTE was introduced. The result of the real-time control showed that the TCSCCFTE performs high accuracy, robust and high reliability.