目的　研究概率神经网络模型 ,并应用于故障诊断 .方法　对基于概率统计思想和 Bayes分类规则的概率神经网络模型、网络结构、算法及其特点进行分析 ,利用其进行故障诊断 ,并提出一种优化估计平滑因子的方法 .结果　概率神经网络可很好地诊断自行火炮发动机进行中油路和气路的故障 .结论　概率神经网络在模式识别和故障诊断领域中可取得良好地应用效果 Objective To study the probabilistic neural network model and apply it to the fault diagnosis.Methods The probabilistic neural network model, network structure, algorithm and its characteristics based on the idea of ​​probability and statistics and Bayes classification rule are analyzed and used to diagnose the fault, and an optimization Method to estimate the smoothing factor.Results Probabilistic neural network can diagnose the failure of self-propelled artillery engine for medium oil circuit and gas circuit well.Conclusion Probabilistic neural network can be applied well in the field of pattern recognition and fault diagnosis