基于Matlab的改进BP神经网络及其应用

Application of Improved BP Neural Network Based on Matlab

讨论了BP神经网络及几种改进BP神经网络性能的算法.用Matlab编制出相应的BP神经网络算法程序,并将其用于处理紫外分光光度法同时测定Fe,Co,Ni,Cu,V的实验数据,结果表明L-M法收敛最快,精度最高.

This paper talks about some improved BP algorithms. We use some programs based on Matlab to deal with the problem of multi-component simultaneous determination, on this basis, choose an optimum one, and then test. It shows that the L-M algorithm is the best.