复合故障下风电齿轮箱声音信号耦合调制模型辨识与故障诊断

Identification and fault diagnosis of sound signal coupling modulation model of wind power gearbox under compound fault

针对两级行星一级平行齿轮箱不同位置受损导致的复合故障,提出一种声音信号耦合调制模型,以辅助专家进行故障诊断。当风电齿轮箱发生复合故障时,其特征频率会以调幅和调频的形式影响不同轮系的啮合频率,为此,本文提出了复合故障下风电齿轮箱声音信号幅值耦合调制模型;利用模型参数辨识思路,确定所提耦合调制模型中不同轮系的调幅系数,并通过构建边带能量比指标,用于评价辨识效果;最后,利用声音信号耦合调制模型的重构谱,确定复合故障位置,实现具有辅助性质的故障诊断。实验与现场数据分析表明:用于评价辨识结果的边带能量比指标分别为0.948,0.972,0.977和0.9643,有效说明了模型辨识的有效性,为齿轮箱复合故障自动诊断奠定了基础。

To assist experts in fault diagnosis,a sound signal coupling modulation model is proposed for compound faults caused by damage at different positions in a two-stage planetary gear system with a single-stage parallel gearbox.When compound faults occur in a wind turbine gearbox,their characteristic frequencies affect the meshing frequencies of different gear stages in the form of amplitude modulation and frequency modulation.Therefore,this paper proposes a coupling modulation model for the amplitude of sound signals in the wind turbine gearbox under compound faults.By utilizing a parameter identification approach,the modulation coefficients for different gear stages in the proposed coupling modulation model are determined.An energy ratio of sidebands index is constructed to evaluate the effectiveness of the identification.Finally,the reconstructed spectrum of the sound signal coupling modulation model is used to determine the location of compound faults,achieving auxiliary fault diagnosis.Experimental and field data analysis show that the sideband energy ratio indicators for evaluating the identification results are 0.948,0.972,0.977,and 0.9643,effectively.These results effectively demonstrate the validity of the model identification,laying a foundation for automatic diagnosis of gearbox compound faults.