The fault signal of the gearbox is the complex signal of the non-stationary and nonlinear characteristics

by using the multifractal detrended fluctuation analysis（MF-DFA）and the support vector machine（SVM）to diagnose the fault of the gearbox.The multifractal detrended fluctuation analysis is employed to the extract multifractal singularity index of the gearbox fault

largest singular index

minimum singular index

the width of the singular spectrum

the singular index of extreme value point and so on

and this four fractal parameters are as characteristic parameter.And then the sample of the gearbox different fault state is built

by using the new method of support vector machine

the gearbox fault diagnosis and identification is achieved.Through the study

a new method for mechanical fault diagnosis is provided through the combination of this two methods.There is of great significance for the fault diagnosis and identification of gearbox.