Zhao Guo, Wang Hui, Liu Gang, Wang Zhiqiang
Key Lab of Modern Precision Agriculture System Integration Research, Ministry of Education of China, China Agricultural University, Beijing 100083, China.
Key Lab of Agricultural Information Acquisition Technology, Ministry of Agricultural of China, China Agricultural University, Beijing 100083, China.
Sensors (Basel). 2016 Sep 21;16(9):1540. doi: 10.3390/s16091540.
An easy, but effective, method has been proposed to detect and quantify the Pb(II) in the presence of Cd(II) based on a Bi/glassy carbon electrode (Bi/GCE) with the combination of a back propagation artificial neural network (BP-ANN) and square wave anodic stripping voltammetry (SWASV) without further electrode modification. The effects of Cd(II) in different concentrations on stripping responses of Pb(II) was studied. The results indicate that the presence of Cd(II) will reduce the prediction precision of a direct calibration model. Therefore, a two-input and one-output BP-ANN was built for the optimization of a stripping voltammetric sensor, which considering the combined effects of Cd(II) and Pb(II) on the SWASV detection of Pb(II) and establishing the nonlinear relationship between the stripping peak currents of Pb(II) and Cd(II) and the concentration of Pb(II). The key parameters of the BP-ANN and the factors affecting the SWASV detection of Pb(II) were optimized. The prediction performance of direct calibration model and BP-ANN model were tested with regard to the mean absolute error (MAE), root mean square error (RMSE), average relative error (ARE), and correlation coefficient. The results proved that the BP-ANN model exhibited higher prediction accuracy than the direct calibration model. Finally, a real samples analysis was performed to determine trace Pb(II) in some soil specimens with satisfactory results.
提出了一种简单但有效的方法，用于在镉（II）存在的情况下检测和定量铅（II），该方法基于铋/玻碳电极（Bi/GCE），结合了反向传播人工神经网络（BP-ANN）和方波阳极溶出伏安法（SWASV），无需对电极进行进一步修饰。研究了不同浓度的镉（II）对铅（II）溶出响应的影响。结果表明，镉（II）的存在会降低直接校准模型的预测精度。因此，构建了一个双输入单输出的BP-ANN，用于优化溶出伏安传感器，该传感器考虑了镉（II）和铅（II）对铅（II）的SWASV检测的综合影响，并建立了铅（II）和镉（II）的溶出峰电流与铅（II）浓度之间的非线性关系。对BP-ANN的关键参数和影响铅（II）的SWASV检测的因素进行了优化。通过平均绝对误差（MAE）、均方根误差（RMSE）、平均相对误差（ARE）和相关系数对直接校准模型和BP-ANN模型的预测性能进行了测试。结果证明，BP-ANN模型比直接校准模型具有更高的预测精度。最后，对一些土壤样品中的痕量铅（II）进行了实际样品分析，结果令人满意。
