灰色理论PNN和D-S集成的汽轮机组故障诊断

汽轮机组发生故障的原因多且复杂,若是仅采用单一的故障诊断方法进行故障诊断,可能会存在误诊或漏诊的问题.为了解决此类问题,同时为了提高故障诊断的可靠性,所以引入D-S理论进行融合故障诊断.针对汽轮发电机组故障的特点,提出了一种基于灰色理论-PNN(概率神经网络)和D-S理论集成的故障融合诊断策略.该策略先采用灰色理论和P...

Integrated fault diagnosis for turbine generator unit using D-S theory based on gray theory and probabilistic neural network

Because the fault cause of turbine generator unit is complex,if a single fault diagnosis method is only used for fault diagnosis,then there may be misdiagnosed and missed diagnosis.In order to solve such problems and im-prove the reliability of fault diagnosis,the D-S evidential theory is introduced into the fusion fault diagnosis.For the characteristics of the turbine generator unit,an integrated fault diagnosis method using gray theory-probabilistic neural network（PNN） and D-S evidential theory was presented.In this method,the results of preliminary diagnosis of fault infor-mation with gray theory and PNN network constitute the evidence firstly,and then,the evidences are fused under the D-S evidential theory and diagnostic results is obtained.Compared with a single diagnosis,the simulation results show that the proposed method can effectively diagnose the vibration fault of turbine generator,can avoid misdiagnosis phenomenon,and has good application prospects.