摘要
针对齿轮故障信号的能量所引起的变化会淹没在常规振动与噪声之中,用传统的信号处理方法不易提取故障特征,给齿轮的故障诊断带来很大困难这一事实,本文描述了用于从振动信号中提取故障信息的小波包和用于识别故障类型的BP网络,研究了BP网络故障模式识别与小波包故障特征提取结合在一起对齿轮故障进行诊断的方法。研究结果表明该方法可以成功地用于齿轮常见故障的识别和诊断。
In view of the fact that the change caused by the energy of the fault signal of gears is drowned by the normal vibration or noise,and the fault characteristics are not easily extracted by means of the traditional signal-processing method,this paper describes the wavelet packet used in extraction of fault information from the gear vibration and the BP neural networks used in identification of fault types, and studies a kind of diagnosis method of gear faults which combines BP neural networks with wavelet packet. The results of the study show that this method can successfully be applied to the identification and diagnosis of gear faults.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2002年第5期659-660,共2页
Journal of Liaoning Technical University (Natural Science)
