利用三维荧光技术进行水质监测对干旱区绿洲河流水质的有效管理具有重要的意义。以三维荧光技术为手段,以艾比湖流域地表水为研究对象,结合平行因子(PARAFAC)法和自组织特征映射神经网络(SOM)方法,探讨了艾比湖流域地表水溶解性有机质的三维荧光特征及其与地表水水质指标之间的关系。通过PARAFAC法,有效提取了艾比湖流域地表水样中的4种荧光组分,C1荧光峰对应物质为紫外区类富里酸,C2荧光峰对应物质为类富里酸,C3包括2个峰C3(T1)和C3(T2),其中C3(T1)荧光峰对应物质为类蛋白,C3(T2)荧光峰对应物质为类腐殖酸,C4荧光峰对应物质为类腐殖质。经SOM训练,在不同聚类层中探讨水质参数分布情况,水质状况由差到好的顺序依次为博河上游、精河绿洲、乌苏周边农田、艾比湖周边。在艾比湖流域丰水期,酸碱度(pH)、电导率(EC)、溶解氧(DO)、化学需氧量(COD)和五日生化需氧量(BOD5)与水样的三维荧光峰具有较为显著的相关性,而总磷(TP)、总氮(TN)及氨氮(NH+3-N)与各荧光峰相关性较弱。分别建立pH、EC、DO、COD及BOD5与各荧光组分间的多元线性回归方程,求得相关系数R分别为0.579、0.632、0.502、0.762和0.785,可以在一定程度上利用各荧光组分模拟水质参数的变化情况。在利用PARAFAC探讨地表水荧光特征的基础上,SOM网络作为一种有效的水体荧光光谱分析工具,可为干旱区水质监测和河流水质污染治理提供科学依据。 The use of three-dimensional fluorescence for water quality monitoring is of great significance to the effective management of oasis river water quality in arid areas. Taking the three-dimensional fluorescence technique as the research object and the surface water in the Aibi Lake basin as the research object, the parallel factor (PARAFAC) method and the self-organizing feature mapping neural network (SOM) method were used to investigate the three-dimensional Fluorescence Characteristics and Its Relationship with Surface Water Quality Index. PARAFAC method was used to extract four kinds of fluorescence components in surface water samples of Lake Aibi. The corresponding fluorescence peak of C1 is the fulvic acid in UV region, the corresponding species of C2 fluorescence peak is the type of fulvic acid, and the C3 includes two peaks of C3 (T1) and C3 (T2). The corresponding material of C3 (T1) fluorescence peak is a kind of protein, the corresponding material of C3 (T2) fluorescence peak is humic acid, and the corresponding material of C4 fluorescence peak is humus-like. After SOM training, the distribution of water quality parameters was explored in different cluster strata. The order of water quality from poor to good was the upper reaches of Bo River, Jinghe Oasis, the farmland around Wusu and the surroundings of Lake Aibi. During the wet season in Lake Aibi, the three-dimensional fluorescence peak (pH), conductivity (EC), dissolved oxygen (DO), chemical oxygen demand (COD) and biochemical oxygen demand (BOD5) (TP), total nitrogen (TN), and ammonia nitrogen (NH + 3-N), and the fluorescence peaks were relatively weak. The multiple linear regression equations of pH, EC, DO, COD, BOD5 and the fluorescence components were established respectively, and the correlation coefficients R were 0.579, 0.632, 0.502, 0.762 and 0.785 respectively, which could make use of the fluorescence components Simulate changes in water quality parameters. Based on the use of PARAFAC to explore the characteristics of surface water fluorescence, SOM network, as an effective tool for analyzing water fluorescence spectrum, can provide a scientific basis for water quality monitoring and river water pollution control in arid areas.