主成分分析法-BP神经网络算法用于电位滴定法测定有机酸

Potentio-Titrimetric Determination of Organic Acids with the Algorithms of PCA and BP-ANN

采用电位滴定法测定混合液中3种以上的多元弱酸,主成分分析法处理了电位滴定的曲线数据,用BP神经网络进行有机酸预测计算。建立了滴定时滴定剂和混合液中有机酸浓度之间关系的模型,实现了不经分离直接测定溶液中几种小分子多元弱酸。结果表明:该神经网络计算出有机酸的预测值与实际值相对误差不大于5.0%。

In the potentio-titrimetric determination of more than 3 weak organic polyacids in mixed solutions,the algorithm of principal component analysis（PCA） was applied to the data treatment of titration curves and the algorithm of back-propagation-artificial neural network（BP-ANN） was applied to presumptive calculation of the organic acids.Model showing the relationship between the amount of titrant consumed and the concentration of organic acids in the mixed solution was established.By the proposed method,weak organ polyacids with small relative molecular masses could be determined in mixed solutions without previous separation.The relative deviations between the pridicted values and the actual contents of the organic acids were found to be ≤5.0%.