Silva Reyes, Anthony

Profile Picture

Filters

20091
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Publication
    Data-driven life prediction model for bearing failure
    (2009) Silva Reyes, Anthony; Goyal, Vijay K.; College of Engineering; Jia, Yi; Valentín Rullán, Ricky; Department of Mechanical Engineering; Steinberg, Lev
    Throughout this work, a prognostic tool for the prediction of the remaining useful life of bearings using experimental data based on vibration data analysis is developed using MATLAB®. The experimental data used us that from NASA Prognostics Repository, where three run-to-failure tests we performed under normal load conditions and data was measured using accelerometers. Our tool is tested using experimental data from the second test in order to confirm a failure on the outer race and perform a remaining useful life prediction. The life prediction is achieved by monitoring the energy level of frequency domain features known as Bearing Fault Frequencies and trending several condition indicators such as Kurtosis, RMS and Power of the interested frequency. Since the damage occurred in the outer race, the Ball Pass Outer Raceway Frequency is selected to perform the life prediction. A set of MATLAB® algorithms are developed to calculate the bearing fault frequencies, trend key condition indicators and perform a life prediction. The remaining useful life prediction is based on the evolution of a selected degradation signal. The selected degradation signal is the RMS vibration level over time. In this approach, instead of calculate the RMS vibration across the entire spectrum; it is calculated around the interested frequency and its first five harmonics using a window of ±15% BPOF. Then the average of these RMS values is calculated and becomes the degradation signal trended over time. Bearing remaining life prediction is achieved based on a predefined failure threshold. Based on the available data, there is an 8.6% error between the actual and predicted bearing failure time.