变负载下的特征提取与基于CHMM算法的齿轮故障诊断

Feature Extraction under Varying Load and Fault Diagnosis Based on CHMM

为了对变负载工况下齿轮故障特征进行有效的提取,提出了基于ARX模型的特征提取方法,并用利用连续隐Markov模型(CHMM)对齿轮的故障进行了识别。实例分析中,对正弦性变化负载的齿轮箱进行了全寿命实验,采用基于ARX模型的方法提取特征,输入到CHMM对故障进行了准确的识别,验证了该方法的有效性。

In order to extract gear fault feature under the condition of various load effectively, a novel feature extraction method based on ARX model was proposed and was identified successfully by using continuous hidden Markov model （ CHMM ） . The performance of the feature extraction method is validated by using full lifetime data obtained from the gearbox operating from a new condition to a breakdown under varying load.