The invention relates to a distribution network material demand prediction method based on a time sequence and hierarchical aggregation, and the method comprises the steps: carrying out the step-by-step prediction of a time sequence prediction model and a multi-dimensional hierarchical structure of any to-be-predicted demand material through a hierarchical aggregation method, and continuously completing the demand prediction of all materials. According to the method, a data mining technology is adopted; implementing high feature classification of goods and materials, profiling material demandrules, establishing prediction model, improving material demand prediction capabilities, material purchasing related work of a material department is supported in an auxiliary mode, material requirements of all project types and power supply units are predicted, accumulated to county companies and finally accumulated to the material requirement scale of city companies, normal production of power enterprises is guaranteed, purchasing and inventory cost is saved to a great extent, and enterprise competitiveness is improved; based on accurate prediction of the material demand quantity, the monthly demand quantity prediction error is smaller than 20%, inventory reduction can be achieved in combination with an information tool, and excessive capital occupation caused by too early excessive purchasing is avoided.