Abstract: In view of the problem that it is difficult to explain the diagnosis process and results of data-driven fault diagnosis method,a method of marine diesel engine fault diagnosis and fault mechanism interpretation based on the classification base of association（CBA） was proposed. Through the GT-Power diesel engine fault simulation experiment,the diesel engine fault database was established,the fault classifier was constructed based on the classification base of association algorithm,and the important rules were extracted based on visualization method to analyze the fault mechanism and the diagnosis process of the model. The results show that the fault recognition rate of the diesel engine fault diagnosis method based on classification base of association is as high as 98.67%,and the extracted rules can better match the principle of thermodynamics related knowledge,which can be used for further research on fault mechanism analysis and fault diagnosis.