In order to more effectively fault diagnosis and identification for aeroengine whole-body vibration, the improved Fuzzy Support Vector Machine (FSVM) and multi-class fuzzy membership with information entropy technique was proposed in this paper.And it compared with the traditional FSVM membership degree analysis methods.The calculation model of multi-class fuzzy membership function.was established based on the traditional FSVM fuzzy membership function. The experiment and examples of aircraft engine overall vibration performance and fault diagnosis and identification verify that the technology of multi-class fuzzy membership with information entropy was very effective.The relationship weights between fault modes and fault causes were determined and the multi-parameter vibration performance analysis model was developed, so that aeroengine overall vibration performance was quantitative analysis and a quantitative reference index was provided for aeroengine vibration control.