摘要
介绍了一种用 RBF网络进行主汽温系统辨识的一种方法 ,采用串—并联型的辨识结构 ,其输入为待辨识对象的输入和输出 ,导师信号为待辨识对象的输出 ,隐层的结构函数采用高斯函数 ,训练时的输入信号采用阶跃信号与白噪声信号的叠加信号 ,仿真结果表明 。
The paper introduces a method of main steam temperature recognition by using RBF network.The structure of recognition is series parallel structure,and the inputs of RBF are the inputs and outputs of the object.The teacher signal is the output of the object and the function of hidden layer is Gauss function.The training signal is the overlaying signal of step signal and white noise.The simulation shows that the method can recognize the system perfectly.
出处
《电力系统及其自动化学报》
CSCD
2003年第4期74-75,100,共3页
Proceedings of the CSU-EPSA
