用BP网络逼近某曲线函数，在神经网络中邮广泛的应用 是系统辨识的基础。



通过反馈来不断的调整系数，限制了最大循环次数 防止进入死循环，当然如果想要得到更精确的系数可以增大循环次数。

-Using BP network function approximation to a curve, mail is widely used in neural network is the basis of system identification.

Through feedback to constantly adjusted factor, limit the maximum cycle times to prevent into dead circulation, and of course if you want to get a more accurate coefficients can increase the cycles.