描述了基于神经网络的康明斯发动机气缸工作状态在线监测的方法。试验结果表明BP神经网络能对从康明斯发动机振动信号中抽取的样本数据给出正确分类，从而对特定气缸工作状态进行有效地识别。此外本文还给出了两种自组织神经网络（SON）分类结果与BP分类结果的比较 This paper describes a method of online monitoring cylinder Cummins engine based on neural network. The experimental results show that the BP neural network can correctly classify the sample data extracted from the Cummins engine vibration signals so as to effectively identify the working state of a specific cylinder. In addition, this paper also gives a comparison between two kinds of self-organizing neural network (SON) classification results and BP classification results