论文部分内容阅读
针对水中存在碘离子、溴离子等干扰物时硝酸根浓度的测定问题,本论文将最小二乘支持向量回归机应用于混合溶液的光谱数据分析并建立起硝酸根、碘离子和溴离子含量的预测模型。实验结果表明:经过本文算法训练后的模型可以较准确的预测混合溶液中硝酸根的含量,基本消除了碘离子、溴离子对测定的干扰。本方法与传统方法相比较,无需化学和物理方法分离,精度高,速度快,适用于水质在线监测系统中对硝酸根含量的测定。
In order to solve the problem of determination of nitrate concentration in the presence of iodide and bromide ions in water, the least square support vector regression (LS-SVM) was applied to the spectral data analysis of mixed solution and to establish the content of nitrate, bromide and iodide Predictive model. The experimental results show that the model trained by this algorithm can predict the nitrate content in the mixed solution more accurately and eliminate the interference of iodide and bromide ions. Compared with the traditional method, the method has no need of separation by chemical and physical methods, has high precision and high speed, and is suitable for the determination of nitrate content in an on-line water quality monitoring system.