【摘要】 应用偏最小二乘-神经网络直接解析硝酸根和亚硝酸根的紫外光谱,不经分离紫外吸光光度法同时测定硝酸根和亚硝酸根。在BP算法上,引用改进的误差传递函数,并采用均匀试验设计法确定了最佳网络运行参数。用于水样中硝酸根和亚硝酸根的同时测定,回收率分别为99.0%,105.0%。更多还原

【Abstract】 A new algorithm of neural network combining with partial least squares method was proposed.The method was applied to simultaneous determination of nitrate and nitrite by ultraviolet spectrophotometry without separation.On the basis of the classic BP algorithm,the error transfer function was improved.The best parameters of network were determined by uniform design.The algorithm was applied to the determination of nitrate and nitrite in water sample with satisfactory results.The mean recoveries are 99.0% and 105.0% for nitrate and nitrite(n=5),respectively.更多还原