Condition assessment of power transformers is an important approach to providing decision-making for condition maintenance and improving the operation reliability of power systems. The power transformer’s operation condition is related to quite a few factors， part of which are vague， making its condition estimation a tough job. A hierarchical grey evaluation method is then introduced to assess the condition of the power transformer. Two models consisting of the abrupt-change condition assessment model and gradual-change condition assessment model are developed via a hierarchical grey relation analysis to evaluate the transformer’s operating condition. The appropriate indices of condition assessment are selected and quantified in two models， respectively. Its variable weights are determined by modifying the constant weight obtained by grey relation analysis so as to improve the accuracy of condition assessment. The results of an example analysis indicate that the hierarchical grey evaluation method proposed is feasible and effective.