BP神经网络用于安痛定注射液中两组分的测定

Determination of two ingredients in antoding injection by the back prepagation nerve network

目的：将人工神经网络用于安痛定注射液紫外光谱的定量分析。方法：网络结构为三层，输入节点数为10，输出节点数为2，隐含节点数为5。算法为Levenberg-marqardt优化方法。结果：氨基比林和安替比林的平均回收率分别为99.2%和100.3%，相对标准差分别为1.9%和1.7%(n=9)。与地方标准方法比较，两者测定结果无显著差异。结论：本法可用于安痛定注射液中氨基比林和安替比林的测定。

Objective: Artificial nerve network was used in the quantitative analysis of antonding injection UV spectrum. Method: Nerve network consisting of three layers of node were trained by using the back propagation learning rule. the numbers of the input node, output node and hidden node were 10, 2 and 5 respectively. Results: The mean recoveries and relative standard deviations of aminopyrin and antipyrine were 99.2%, 100.3% and 1.9%, 1.7% (n=9).Compared with standard method, no notable difference were found. Conclusion: The proposed method can be used to determinate aminoprin and antipyrine in antoding injection.