Abstract: An aeroengine fault diagnosis method based on support vector machines is presented.After extracting fault symptoms via proper equipments or methods,support vector machines were adopted to set up correlation between fault symptoms and fault patterns and to realize the classification of faults. Support vector machines are specially designed for good generalization ability under the condition that only small samples set is available.Applying this method to a certain engine,the diagnostic results are satisfactory,and it is referable to develop diagnostic system for newly-developed engines.