Recognition of Damage Degree of Tooth Root Crack based on PCA and Grey Relational
In the aspect of root crack fault detection， the severity of damage can be compared by cepstrum analysis， but it is impossible to quantify the extent of damage. In order to realize quantitative detection of damage degree， the method of Principal Component Analysis （PCA） and Grey Relational Analysis...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of Mechanical Transmission
2020-01-01
|
| Series: | Jixie chuandong |
| Subjects: | |
| Online Access: | http://www.jxcd.net.cn/thesisDetails#10.16578/j.issn.1004.2539.2020.09.021 |
| Summary: | In the aspect of root crack fault detection， the severity of damage can be compared by cepstrum analysis， but it is impossible to quantify the extent of damage. In order to realize quantitative detection of damage degree， the method of Principal Component Analysis （PCA） and Grey Relational Analysis （GRA） are combined to use. Firstly， the energy method is used to calculate the time-varying meshing stiffness of gear pairs with different root cracks and analyze the response of different damage degrees. Then， statistical indicators of damage detection are also combined to carry out quantitative detection. Through PCA algorithm， the dimensional reduction of the multidimensional statistical indicators of damage detection is optimized. The relational degree between the target sequence to be examined and each comparison state sequence is calculated. The degree of crack damage is characterized by relational degree. On the basis of theoretical simulation， experimental verification is carried out. The results show that cepstrum analysis can effectively identify the root crack fault. The higher the damage degree， the higher the peak amplitude of cepstrum. The relational degree of PCA and GRA combined algorithm is larger and the discrimination degree is more obvious than that of direct GRA calculation. The damage degree of the target will can be effectively quantified， the theoretical basis for quantitative identification of tooth root crack is provided. |
|---|---|
| ISSN: | 1004-2539 |