摘要
目的:探讨应用紫外分光光度法同时测定混合体系中苯和甲苯的可行性。方法:根据反向传输(BP)算法,应用三层人工神经网络(ANN)原理,对紫外光谱严重重叠的苯和甲苯的混和体系进行测定。结果:苯和甲苯的回收率分别为98.7%、99.4%,测定结果的相对标准偏差分别为2.0%、2.8%。结论:ANN辅助紫外光谱法具有训练速度快、测定周期短、预测结果准确度高等优点。
Objective: By using spectrophotography, to find out feasible determination's method of the mixture of benzene and methylbenzene at the same time. Methods:By means of artificial neural network and back-propagation training algorithm, the two components of benzene, methylbenzene in which the spectra overlapped were determined simultaneously. Results: The mean recovery of benzene and methylbenzene were 98.7%, 99.4% respectively. The RSD of the results were 2.0% and 2.8%. Conclusion: This method has many advantage such as high training speed, short assay cycle and the high accuracy of predict outcome.
出处
《新疆医科大学学报》
CAS
2005年第6期519-520,共2页
Journal of Xinjiang Medical University
关键词
人工神经网络
紫外光谱
苯
甲苯
artificial neural network
ultraviolet spectrum
back-propagation algorithm
benzene
methylbenzene
