提出了结合主成分分析法(PCA)的可拓神经网络算法，并且将其应用于柴油、煤油、汽油的三维荧光光谱分类识别中。实验结果表明，相比传统的BP神经网络算法，该算法迭代数下降了80步，识别率由89%提高到93%，体现了结合算法的高识别率和高效性。

A combination of principal component analysis (PCA) and extension neural network algorithms has been employed for the classification and recognition of the threedimensional fluorescence spectra of diesel, kerosene and gasoline. Compared to the traditional BP neural network algorithm, there was a decrease of 80 in the number of iterations of the algorithm required and the recognition ratio increased from 89% to 93%, showing the high recognition ratio and efficiency of the combined algorithm.