遗传神经网络-紫外分光光度法同时测定苯酚和邻苯二酚  

Simultaneous determination of phenol and pyrocatechol by genetic neural network-UV spectrophotometry

作　　者：张成燕[1] 马卫兴[1] 毛应明[1] 周开静 

机构地区：[1]淮海工学院化学工程系,江苏连云港222005

出　　处：《计算机与应用化学》2009年第10期1300-1302,共3页Computers and Applied Chemistry

基　　金：江苏省高校自然科学研究计划资助项目(05KJB150003);江苏省海洋生物技术重点建设实验室基金课题(2005HS010)

摘　　要：研究吸收光谱重叠严重的苯酚和邻苯二酚的两组分体系,针对BP神经网络易陷入局部极小等缺陷,将遗传算法与BP神经网络相结合,用遗传算法优化神经网络的初始权值和阀值,由神经网络输出的误差构造适应度函数,建立遗传神经网络算法,用紫外分光光度法同时测定混合的苯酚和邻苯二酚,预测集样品的相对平均误差分别为0.818%和0.366%,对水样的加标回收率分别为104.7%和102.9%。The two components system of phenol and pyrocatechol was studied by UV spectrophotometry with serious overlapping peaks. Considering some defects of back-propagation neural network （BP） , the model was set up by optimization of initial weights and thresholds of neural network using genetic algorithm and designing fitness function by output error. The contents of phenol and pyrocatechol were determined simultaneously by GA-BP-ANN model and ultraviolet spectrophotometry. For phenol and pyrocatechol, the relative mean errors in the prediction set were 0. 818% and 0. 366% and the recovery rate of water sample were 104. 7% and 102. 1%.

关 键 词：苯酚 邻苯二酚 遗传神经网络 紫外分光光度法 

分 类 号：TQ015.9[化学工程] O6-39[理学—化学]