• 关键词：近红外光谱  废水  COD  光程

• 基金项目：福建省自然科学基金(NoZ0513014);浙江省教育厅项目(NoG20060346)

• 何金成
• 1. 浙江大学生物系统工程与食品科学学院, 杭州 310029; 2. 福建农林大学机电工程学院, 福州 350002

• 杨祥龙
• 浙江大学生物系统工程与食品科学学院, 杭州 310029

• 王立人
• 浙江大学生物系统工程与食品科学学院, 杭州 310029

• 泮进明
• 浙江大学生物系统工程与食品科学学院, 杭州 310029

• 摘要：收集了城市生活污水处理厂和养殖废水200个样品,以4cm^-1的分辨率,分别用2mm和10mm样品池,用透射法采集800～2500nm谱区的光谱数据,采用偏最小二乘法(PLS)回归建立了废水的生化需氧量(BOD₅)、化学需氧量(COD)和pH的预测模型.该模型的BOD₅、COD和pH的预测值与参考值的校正相关系数分别为0.763、0.975、0.966,校正标准误差分别为27.7mg·L^-1、19.9mg·L^-1、0.165,预测相关系数分别为0.6939、0.945、0.927,预测标准误差分别为37.9mg·L^-1、29.4mg·L^-1、0.208.实验结果表明,近红外光谱技术可成为特定废水多指标快速检测的新手段.

• Abstract：The aim of this study is to develop a multi-index water quality characterization protocol using near infrared(NIR) spectroscopy.Two hundred wastewater samples were collected from a wastewater treatment plant,a fishpond and a hoggery.The spectral data were collected in the region of 800～2500nm using InGaAs detectors at a 4cm^-1 resolution using cell pathlengths of 2mm and 10mm.Apartial least squares(PLS) model for the determination of chemical oxygen demand(COD),biochemical oxygen demand(BOD₅) and pHwas developed.The calibration coefficient and standard error were 0.975 and 19.9 mg·L^-1 for COD;0.759 and 32.7 mg·L^-1 for BOD₅;0.966 and 0.165 for pH.The correlation coefficient and the standard error of prediction were 0.945 and 29.4 mg·L^-1 for COD;0.6939 and 37.9 mg·L^-1 for BOD₅;and 0.927 and 0.208 for pH.These results indicate that NIRspectroscopy has great potential for application in the determination of multi-index wastewater quality parameters.

• Key words：near-Infrared spectroscopy  wastewater  chemical oxygen demand(COD)  biochemical oxygen demand(BOD₅)  pH

• 
  摘要点击次数： 2575 全文下载次数： 3059