An Application of Maximum Entropy Method to Fault Diagnosis of Rolling Bearings
Xu Yu · Mechanical Science and Technology · 2001
AR model is built by maximum entropy method on bearing running condition, and time series model of fault and normal condition is computed. Maximum entropy analysis is carried on time series which contains signal. By analysis, the qualitative and quantitative diagnosis is processed, contemporaneously compared with FFT (power spectrum), spectrum produced by maximum entropy have many advantages: Limited valvula, standing out main peak, high frequency resolution ratio. Diagnosis information is improved.