中文摘要：

利用交替三线性分解（ATLD）、交替拟合残差（AFR）和自加权交替三线性分解（SWATLD）等3种二阶校正算法分别对三维荧光光谱数据进行解析，实现了香蕉中的双苯三唑醇含量的直接快速定量测定。当被分析体系的组分数取2时，ATLD、AFR和SWATLD获得的平均回收率分别为（98．2±2．6）％、（97．6±2．1）％和（91．5±3．5）％。另外，采用椭圆置信区间测试（EJCR）和品质因子，如灵敏度（SEN）、选择性（SEL）和检出限（LOD）评估了3种算法的准确性。实验结果表明：3种算法均能成功用于直接分析香蕉中双苯三唑醇的含量。

英文摘要：

A novel approach was developed for direct quantitative analysis of bitertanol in banana by the incorporation of three-dimensional fluorescence with second-order calibration methodologies based on alternating trilinear decomposition （ATLD）, alternating fitting residue （AFR） and self-weighted alternating trilinear decomposition（SWATLD） algorithms respectively. The average recoveries of bitertanol in banana extract by using ATLD, AFR and SWATLD with an estimated component number of two were （98.2± 2. 6） % , （97.6 ± 2. 1）% and（91.5 ±3.5）% , respectively. Moreover, the accuracy of these three algorithms was also evaluated through elliptical joint confidence region tests（EJCR） as well as figures of merit, such as sensitivity（SEN）, selectivity（SEL） and limit of detection（LOD）. The experimental results demonstrate that these algorithms have been successfully applied to determine the content of bitertanol in banana extract.