In view of the fact that a marine gas turbine is prone to compressor scaling, abnormal axial force, temperature field distortion and even abnormal blade wear when it runs in the marine environment for a long time, this paper aims to use the big data analysis method to study the degradation process of the gas turbine in use. Based on the traditional experience analysis method, this paper puts forward a method of establishing degradation evaluation index based on correlation analysis, establishes a degradation evaluation model based on support vector description, introduces the concept of degradation degree, and analyzes the degradation of a certain gas turbine through experiments, and achieves good results.