介绍了前馈网络的结构,BP算法以及算法的推导过程,并对算法做了一些改进.然后将该算法应用于抽油机减速箱的状态识别,给出了识别结果.实验表明,基于BP算法的前馈网络在识别的精度和自学习方面都优于传统方法. The structure of the feedforward network, the BP algorithm and the derivation process of the algorithm are introduced, and some improvements are made to the algorithm.And then, the algorithm is applied to the state recognition of the gearbox of the pumping unit, and the recognition result is given.Experiments show that based on BP The algorithm’s feedforward network is better than the traditional method in recognition accuracy and self-learning.