精确算法径向基神经网络在同时测定Cu,Zn,Pb中的应用

用非线性的方法来解决多组分金属离子体系同时测定的问题。用NATLAB编写了前向基多层神经网络(BP)和精确径向基神经网络(RBFE)，并比较了它们在校正时的情况，发现在此体系中RBFE网络优于BP网络，因此将RBFE网络应用于3组分金属离子Cu、Zn、Pb的同时测定。使用络合滴定的方法，在待测溶液中不加缓冲液而直接用EGTA滴定，从而得到一系列的滴定曲线，通过对这些曲线的分析，以达到简化分析程序的目的。

RBFE Applied to Simultaneous Determination of Cu, Zn, Pb by Complexometric Titration Method

A non-linearity method was used to resolve the problem for simultaneous measuration of concentration of metal ion mixtures.Here we wrote the a multilayer feedforword net with the backpropagation learning algorithm(BP)and a neural network based on exact radial basis functions (RBFE)with MATLAB v5.3,and compared the circumstance for them to correct.We discovers that the RBFE network of system inside is better than BP,therefore RBFE network is applied to simultaneous measurement of three transition metal ions such as Cu~2+,Zn~2+,and Pb~2+.A series of titration curves were made by complexometric titration whth ethylene glycol-bis(2-aminoethylether)-N,N,N,N,-tetraacetic acuid(EGTA)in case of no buffer solution ,and a simplified analysis procedure was got by analyzing these curves.