Application of general fractal dimension to rolling bearing diagnosis

The multi-fractal theory is applied for proposing the calculation formula of general fractal dimension according to the characteristic of rolling bearing vibration signal. The general dimension of measured non-stationary time-domain signal is calculated and analysed, the box dimension, the information dimension and the correlation dimension are obtained. Combined general dimension sequence value and mathematics method, the fractal diagnosis classification theory is presented. The fault pattern is recognized and classified by the utilizing theorem of the general dimension correlation coefficient method and general dimension series single value approach method. The theory and experiment analysis show that the general dimension correlation coefficient method is a preferable method in fault pattern recognition for the few samples and high recognition rate. The working conditions is indicated by the general fractal dimensions directly. It is effective to recognize the fault pattern without disassembling the bearing according to the general dimensions. The simple and feasible method is a new approach to the rolling bearings fault diagnosis.