Comparison of roughness level in ball bearing using statistical analysis of vibration signal

Cambow, Rameshwar; Singh, Manpreet

doi:10.1016/j.matpr.2021.05.001

Cited by 1 publication

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The usage of statistical parameters in vibration signal analysis continued to be a preferred choice of many authors because of its simplicity, precision, reliability, and quicker computation [36][37][38][39][40][41][42][43][44][45][46][47][48]. Moreover, statistical parameters have also been used to extract information about the level of surface roughness [23][24][25][26], which is of prime concern in this research work.…”

Section: Theoretical Backgroundmentioning

confidence: 99%

“…Because of its capacity to process signals on a logarithmic scale, it may serve as a significant criterion for representing the diverse features of coating in addition to its many uses. Mathematically it is expressed as [47]:…”

Section: Variancementioning

confidence: 99%

“…Shannon entropy is another important statistical parameter and widely used in diagnostic and monitoring analysis of bearing [25,[43][44][45][46][47]. And also used in the study carried out to extract features of vibration signal for the monitoring of surface finish in CNC operation [55].…”

Section: Variancementioning

confidence: 99%

See 2 more Smart Citations

Comparative analysis of surface roughness level for bearing coating using VMD of vibration signal

Cambow,

Singh

2024

Phys. Scr.

View full text Add to dashboard Cite

Degradation in surface finish of the material is an important concern in engineering applications. However, most of the conventional techniques available to monitor the surface roughness requires dismantling of the machine element. Though in rotary components like bearing, it is practically not feasible. Therefore, in this work, an on-board technique is proposed to compare the level of surface roughness based on vibration signature. To demonstrate, five different industrial standard coatings (Nickel, Copper, Zinc phosphate (ZnP), Silver, and Black oxide) were carried out on raw five ball bearings (NBC6205). Two sets of coated bearing were prepared, where first set was utilized for experimentation and second for the purpose of measuring surface roughness of bearing surface. These coated bearings were tested at five different RPMs ranging from 300 RPM to 1500 RPM and their vibration signals were recorded. The recorded vibration signals must be having characteristics originated from ball rolling on different level of surface roughness and hence distributed in nature. Further, to target distributed characteristics present in the signal, commonly used statistical parameters for vibration signature analysis (RMS, Crest factor, Variance, Skewness, Kurtosis, Shannon entropy and Log energy) were calculated. Then, correlation of the parameters was checked in relation to the different levels of surface roughness, but no relation was found. The signals were then decomposed into six intrinsic mode functions (IMFs) using Variational Mode Decomposition (VMD) method. Again, same statistical parameters were calculated for these decomposed levels, it has been noted that Shannon entropy have shown correlation to surface roughness in at least one decomposed level between 900 to 1500 RPM with minimum value of chain index as 176.65. Moreover, in this RPM range, responsive frequency bands found to be shifted towards higher side i.e. from 567 to 4590 Hz.

show abstract