铅锌的湿法冶金实时监控需要同时快速测定溶液中的铅,镉,锌,因三者产生的吸收光谱严重重叠,故用支持向量回归算法(support vector regression,简称SVR)作多变量标定,配合吸光光度法同时测定溶液中的Pb~(2+)、Cd~(2+)、Zn~(2+),并将计算结果与线性回归、人工神经网络的计算结果对比。结果表明SVR算法预报误差最小。这说明支持向量回归是一种多变量校正的有用算法。 Lead-zinc hydrometallurgical real-time monitoring needs simultaneous rapid determination of lead, cadmium and zinc in solution, because the absorption spectra produced by the three have serious overlapping spectra. Therefore, support vector regression (SVR) With the simultaneous determination of Pb ~ (2 +), Cd ~ (2 +) and Zn ~ (2+) in solution by spectrophotometry, the calculated results were compared with the results of linear regression and artificial neural network. The results show that SVR algorithm has the smallest prediction error. This shows that support vector regression is a useful algorithm for multivariate calibration.