Recognition Of The Vibration Signals Based On Ica

When the mechanism and architecture have some failures, the interior structure, exterior shape or work condition will have the abnormal representation. It is practically important for the diagnose and maintenance of mechanism and architecture that implement and abnormity detection of abnormal vibratory signals obtained from the surface of the objects.The mechanical device vibration signal contains the plenty operation of equipment information. It is the important origin of diagnosis. But the vibration signal is often not single, because it is influenced by installation position of sensor and uncertainty of the breakdown actual vibration direction when measuring the vibration signal. So other disturbance signals are included in the survey signal inevitably. On the other hand, the new vibration can be produced possibly because of the induction vibration of the breakdown. The above reason has caused the present equipment examination and the breakdown processing accuracy improved difficulty. The independent component analysis opens a new research way for solution above question.In this paper, the application of independent component analysis in vibration signal processing has been studied. The paper consists of following parts:First, the paper reviews systematically the present research situation of independent component analysis in the world. The basic principles and concepts of independent component analysis and some algorithms are introduced.Second, the vibration source recognition apply a function based on the foundation of the fast independent component analysis algorithm, It separated the mixed voice and the mixed vibration source well;Third, as to the number of the mixed signal more than the source, the Singular Value Decomposition is applied, the method solved the number of the source and the noise; the matrix joint-diagonalizing pre-whitening JADE algorithm has been introduced, and the performance of the algorithm has been illustrated by computation simulation experiment. The result display the performance of the algorithm is good.At last, summary some questions and the next steps of development direction which this paper work and need to solve.