The accurate diagnosis of transformer oil-paper insulation state has important guiding significance for the formulation of maintenance strategy. In order to realize the objective scientific evaluation of the insulation state of oil-paper, a multi-feature comprehensive evaluation model of transformer oil-paper insulation state based on combined weight-double base point method is proposed. Firstly, based on the shortcomings of the single feature quantity evaluation, based on the existing time-domain dielectric spectrum data, multiple aging feature quantities are extracted from the recovery voltage polarization spectrum and the extended Debye equivalent circuit respectively, and a standard evaluation index system is established.Secondly, it proposes an index combination weighting model that takes into account the subjective experience of experts and objective differences in data, overcoming the shortcomings of the single weighting method. Finally, the closeness of the object to be evaluated and the ideal point of the standard state is calculated based on the double base point method, and the comprehensive evaluation result of the oil paper insulation state is obtained.The example verification shows that the evaluation model built in this paper can effectively and accurately diagnose the insulation state of the transformer, which can provide a new method for the evaluation of transformer oil-paper insulation state.