过热汽温系统的RBF神经网络控制

RBF Neural Network Control on Super-heated Steam Temperature System

采用RBF神经网络直接构成神经网络控制器,将在线学习和控制相结合,这种方法不需要增加另一个神经网络对系统进行在线辨识,也不需要预先确定神经网络控制器的结构。通过将该方法应用于电厂过热汽温系统的控制进行仿真研究并与常规PID串级控制系统进行比较,结果表明控制系统的性能得到较大的提高。

Using the RBF neural network as neural network controller, combining the on-line learning and controlling, a new neural network control method is presented, and another neural network for on-line system identification and determining the structure of neural network controller a prior is no need. The simulation for super-heated steam temperature control system using presented method is carried out. The results show that the control system performance is better than the conventional cascade control system.