CAJ | 학술논문

行星传动齿轮故障信号频谱中蕴含重要的特征信息，但故障信号湮没大量噪声，通常的谱分析算法难于有效提取故障特征。提出一种多维小波分块阈值降噪方法，通过计算行星轮内环通过频率特征、外环通过频率特征和自转频率特征，得到加速度频谱和位移频谱双频谱，实现对行星轮故障的双谱分析。仿真实验通过建构信号和行星传动轮系多分布点采集故障信号进行测试。结果表明，降噪算法比传统算法性能优越，可以有效诊断齿轮故障，得出诊断结论，有效实现故障状态监测。
행성전동치륜고장신호빈보중온함중요적특정신식，단고장신호인몰대량조성，통상적보분석산법난우유효제취고장특정。제출일충다유소파분괴역치강조방법，통과계산행성륜내배통과빈솔특정、외배통과빈솔특정화자전빈솔특정，득도가속도빈보화위이빈보쌍빈보，실현대행성륜고장적쌍보분석。방진실험통과건구신호화행성전동륜계다분포점채집고장신호진행측시。결과표명，강조산법비전통산법성능우월，가이유효진단치륜고장，득출진단결론，유효실현고장상태감측。
The planetary gear fault signal contains important information, but it also contains lots of noise. Traditional spec-trum method cannot effectively extract the fault characteristic spectrum. An improved block threshold de-noising method with multidimensional wavelet analysis was proposed for noise reduction. Through the calculation of planetary ring through a frequency characteristic, the outer ring frequency characteristics and the rotation frequency characteristics, we got the ac-celeration spectrum and the displacement spectrum as the double spectrum. Simulation was implemented with the real multi point distribution fault signal. Simulation result shows that noise reduction algorithm is superior to traditional algo-rithm, and it can effectively diagnose the gear fault diagnosis with the detail conclusions. It can be applied in the real-time fault monitoring.