简介了人工神经网络（ＡＮＮ）的定义及分类，着重介绍了ＡＮＮ中一种较经典的网络模型－ＢＰ网络模型的网络结构和学习原理。总结了ＡＮＮ在化学领域中的应用，包括谱图分析（质谱、红外、核磁）、药物分子药效判定、蛋白质结构预测等。同时讨论了在使用ＡＮＮ方法时应注意的几个问题，如隐层节点数目的确定、如何防止过拟合等。 This paper introduces the definition and classification of artificial neural network (ANN), and introduces the network structure and learning principle of a more classic network model - BP network model in ANN. The application of ANN in chemistry was summarized, including spectrum analysis (mass spectrometry, infrared and nuclear magnetic resonance), determination of drug molecular potency, prediction of protein structure and so on. At the same time, several problems that should be noticed when using ANN method are discussed, such as how to determine the number of hidden nodes and how to prevent over fitting.