The application of correlation dimension in gearbox condition monitoring

This paper reports a study of correlation dimension in gearbox condition mo nitoring. In contrast to other fault diagnosis methods, such as Fourier spe ctrum analysis? time-frequency analysis, etc., the correlation dimension ca n provide some intrinsic information of an underlying dynamic system recons tructed from measured scalar time series. A three-stage analysis procedure using correlation dimension is presented. Some important influencing factor s relating directly to the computational precision of correlation dimension are discussed. Industrial gearbox vibration signals measured from differen t operating conditions are analyzed using the above method. Results show th at the correlation dimension is able to identify clearly a gearbox-operatin g condition with fatigue crack or broken tooth compared with the normal con dition. (C) 1999 Academic Press.