Optimization of chromatographic conditions in the base on feed forward neural networks with non linear inputs
ZHU Ping 1,GUO Wei qiang 2,XU Miao yun 1,TANG Jin shang 1(1.Dept. of Physics,Zhejiang Univ.,Hangzhou 310028,China; 2.Dept. of Chemistry,Zhejiang Univ.,Hangzhou 310028,China)
In this paper,the optimization of the thermal desorption gas chromatographic experimental parameters is investigated using feed forward neural networks with non linear inputs.It is found that the reflection from experimental parameters to evaluating indices can be described well with the neural networks.When the number of hidden nodes is taken as 6, the times for training networks are almost same for both non linear inputs and linear inputs.
【CateGory Index】： O657