本文重点讨论正向辨识建模，提出了一种基于ＢＰ网络的改进的ＢＰ神经网络（ＩｍｐｒｏｖｅｄＢＰＮｅｕｒａｌＮｅｔｗｏｒｋ简称为ＩＢＰ）用于正向建模的方法。并以非线性的具有滞后的系统为例进行了仿真，对采用常规的ＢＰ网络和ＩＢＰ网络用于正向辨识建模的情况进行了比较，得出了比较理想的结果 This paper focuses on the forward identification modeling, and proposes an improved BP neural network (Improved BP Neural Network referred to as IBP) method for forward modeling based on BP network. A nonlinear system with hysteresis is taken as an example to simulate the situation that the conventional BP network and IBP network are used for forward identification modeling and the ideal result is obtained