Aiming at the major difficulties of OMA method based on vibration signals such as low noise-robust ability,high subjective experience and poor accuracy in identifying parameters and more complex recognition method,a new method of OMA on the basis of QICA is studied.Through the LMS modal experimental data algorithm analysis,the QICA modal parameter identification technology operates simply,and can effectively eliminate noise interference.Finally,regards taking of gearbox modal frequency of the five kinds of working condition recognized by the QICA as characteristic parameter of fault diagnosis,the changes of any kind of model frequency on gearbox fault working condition are compared and analyzed,and the goal of the fault identification is achieved.