基于BP网络的跳汰机松散度软测量建模及应用研究

In this paper，jig loose layer of soft measurement model for the application of research．This question would be a complicated nonlinear problem and could be difficultly solved with the conventional mathematical method．In view of such problem，the BP neural network method was applied to establish a soft-sensing mode of the jig loose layer for the nonlinear dynamic processing capacity and the ability of actually to state the nonlinear relationship between input and output．The variables of v0,h0,hm,T are used as the neural network input．P is taken as the neural network output．Through interactively train and study，the intrinsic relation between input and outputis drawn in the neural network. Experiment results show that the proposed method based on BP neural network has strong capability in dynamic approach and a high application value.