用神经网络-方波溶出伏安法同时检测微量元素

在（pH=1．5）硝酸钾-硝酸溶液中，采用方波溶出伏安法对铜、铅、镉和锌四种金属离子的混合溶液进行信号采集，并用人工神经网络处理方波溶出伏安信号，建立了性能良好的四种金属离子同时测定的神经网络测试模型。用该模型同时对食醋中铜、铅、镉、锌的含量进行了测定，结果表明：人工神经网络能够较好地解决金属离子之间的相互作用和伏安信号干扰问题，食醋样品中此四种金属离子含量的测量结果令人满意。因此，此方法必将在食品分析中有着广阔的应用前景。

Simultaneous Determination of Trace Copper, Lead, Cadmium and Zinc by Means of Artificial Neural Network and Square Wave Stripping Voltammetry

The stripping analysis responses were obtained in the solutions containing varying concentrations of Cu,Pb,Cd,and Zn by Square Wave Stripping Voltammetry in supporting electrolytes of KNO_3/HNO_3(pH1.5).A feed-forward neural network was trained to model the relationship between response and concentrations in the situations of simultaneous determination of the four heavy metals.The model was validated by simultaneous determination of the four heavy metals in vinegar sample.The results show that neural network can be utilized to solve the problems of the complication of stripping voltammetric responses by the interaction of metal ions.The result of quantifying the four heavy metals in vinegar by this method is satisfactory.Therefore,the method will enjoy wide application prospects in food analysis.