[1]刘梓晶,李红岩,王新,等.荧光光谱-KNN算法用于腐殖酸种类识别和定量[J].中国给水排水,2022,38(23):56-62.
　LIUZi-jing,LIHong-yan,WANGXin,et al.Identification and Quantification of Humic Acid Based on K-Nearest Neighbor Algorithm Combined with Fluorescence Spectra[J].China Water & Wastewater,2022,38(23):56-62.
点击复制

荧光光谱-KNN算法用于腐殖酸种类识别和定量

中国给水排水[ISSN:1000-4062/CN:12-1073/TU] 卷: 第38卷 期数: 2022年第23期 页码: 56-62 栏目: 出版日期: 2022-12-01
Title:
Identification and Quantification of Humic Acid Based on K-Nearest Neighbor Algorithm Combined with Fluorescence Spectra
作者:
刘梓晶1,2， 李红岩1， 王新1， 郑蓓1， 黄向阳2， 高迎新1
（1.中国科学院生态环境研究中心 饮用水科学与技术重点实验室，北京 100085；2.长江大学 城市建设学院，湖北 荆州 434000）
Author(s):
LIU　Zi-jing1，2， LI　Hong-yan1， WANG　Xin1， ZHENG　Bei1， HUANG　Xiang-yang2， GAO　Ying-xin1
（1. Key Laboratory of Drinking Water Science and Technology， Research Center for Eco-Environmental Sciences， Chinese Academy of Sciences， Beijing 100085， China； 2. Urban Construction College， Yangtze University， Jingzhou 434000， China）
关键词:
腐殖酸;  稳态瞬态荧光光谱仪;  KNN算法;  饮用水;  水源水
Keywords:
humic acid;  steady/transient fluorescence spectrometer;  K-nearest neighbor algorithm;  drinking water;  source water
摘要:
生活饮用水及水源水中腐殖酸（HA）的种类识别和定量对消毒副产物的控制有重要意义。为此，通过稳态瞬态荧光光谱仪提取腐殖酸的荧光特征值即荧光指数（FI）、生物源指数（BIX）和腐殖化指数（HIX）信息，引入KNN特征值模型，建立了荧光光谱-KNN算法技术识别水中的腐殖酸种类，并在此基础上，利用同种类腐殖酸标准物质的荧光强度与质量浓度的线性关系进行定量研究。结果表明，该方法对水中6种腐殖酸标准物质的检出限范围为0.001~0.002 mg/L、定量限范围为0.003~0.006 mg/L、精密度范围为0.29%~1.80%、准确度范围为1.3%~3.6%，加标回收率范围为86.5%~110%，均满足分析化学在定量方面的相关规定。在定性准确、数据可靠的条件下，利用荧光光谱-KNN算法技术对14种不同来源的生活饮用水及水源水样品进行了腐殖酸种类和含量的调查，结果发现，调查范围内水体中的腐殖酸均与标准物质1R110N属于同一种类，其中水源水中的腐殖酸浓度为1.922~9.157 mg/L，出厂水中的腐殖酸浓度为0.382~2.108 mg/L，水厂对腐殖酸的去除率为44.02%~83.78%，证实了该方法在应用推广方面具有普遍适用性。
Abstract:
It is of great significance to identify and quantify humic acid in drinking water and source water for the control of disinfection byproducts. Therefore, the fluorescence characteristic values such as FI, BIX and HIX information of humic acid were extracted by steady/transient fluorescence spectrometer, and K-nearest neighbor (KNN) algorithm characteristic model was introduced to establish the fluorescence spectra-KNN algorithm to identify the humic acid species in water samples. On this basis, the linear relationship between fluorescence intensity and mass concentration of homogeneous humic acid standard materials was quantified. The detection limits of six humic acid reference substances in water samples were 0.001-0.002 mg/L, the limits of quantification were 0.003-0.006 mg/L, and the precision, accuracy and standard recovery were 0.29%-1.80%, 1.3%-3.6% and 86.5%-110%, respectively. All of them met the quantitative requirements of analytical chemistry. Under the condition of accurate qualification and reliable data, the humic acid types and concentrations of 14 kinds of drinking water and source water samples were analyzed by the fluorescence spectra-KNN algorithm. The humic acid in the water samples under investigation belonged to the same standard substance as 1R110N, among which the humic acid concentration in source water was 1.922-9.157 mg/L, and the humic acid concentration in the treated water was 0.382-2.108 mg/L. The removal rates of humic acid in the water treatment plants were 44.02%-83.78%. The results proved that the method had universal applicability in application and promotion.

相似文献/References:

[1]王轶,陈卫,耿楠楠.基于TiO2/石墨烯的光催化材料降解腐殖酸[J].中国给水排水,2020,36(9):83.
[2]魏永,吴宏.纳滤处理太湖水的膜污染中试研究[J].中国给水排水,2021,37(7):66.
　WEI Yong,WU Hong.A Pilot Study on Membrane Fouling in the Treatment of Taihu Lake Water by Nanofiltration[J].China Water & Wastewater,2021,37(23):66.
[3]王晓云,付爱民.二氧化氯预氧化生成无机消毒副产物的研究[J].中国给水排水,2021,37(23):34.
　WANG Xiao-yun,FU Ai-min.Formation of Inorganic Disinfection By-products during Chlorine Dioxide Pre-oxidation[J].China Water & Wastewater,2021,37(23):34.

更新日期/Last Update: 2022-12-01