Transformer is one of the critical equipment for electric power transmission and distribution, and its safety situation plays a great effect on stability and security level of power system. The material aging and equipment deterioration can both affect the remaining life immensely, and then influence the stability of power system. Therefore, researches on health situation and remaining life of transformers are receiving more attention in academia. The article proposes a transformer life prediction model based on the improved grey forecasting and entropy method. The health index system of the model takes advantage of both hierarchical analysis system and reliability index system, which could overcome the limits of each system. Considering the transformer-life-related factors, such as operating environment, load rate and fault or defects, the model puts forward several modifying factor of health index, which can greatly improve the assessment ability in transformer health situation. The efficiency of the model is successfully verified by examples, and the model can provide useful guidance and suggestion in the decision-making work and has a great potential in the application of transformer life prediction.