Abstract: The quality of distribution network equipment is the basis of the construction of intrinsically safe distribution system, and the sample inspection of delivered equipment is an important means of the quality control of distribution network equipment. Centering on the problem that distribution transformer capacity is hard to be inspected accurately at present, this paper proposes a new capacity inspection method for distribsution transformer based on big data. This paper accumulates a large number of test data of nearly one thousand distribution transformers from inspection work, and analyzes the distribution of volume, weight, DC resistance, no-load loss, load loss and short-circuit impedance of distribution transformer in different capacity levels through the statistical method, and finally makes a capacity assessment process. This method is simple, practical and reliable, and it can be spread and applied in production practice.