论文部分内容阅读
针对旋转设备六种常见故障，采用 Ｂ Ｐ网络，运用模糊理论产生训练样本，根据经验公式和实际运算结果选择网络拓扑结构。经验证，其对单一故障具有很好的识别能力，与传统 Ｂ Ｐ网络的诊断结果相比，提高了诊断精度。
Aiming at six kinds of common faults of rotating equipment, B P network is adopted, fuzzy theory is used to generate training samples, and network topology is selected according to empirical formula and actual operation results. It has been proven that it has a good ability to identify single faults and improves the diagnostic accuracy compared to the diagnostic results of traditional B P networks.