The invention relates to a method for determining electricity demand factors of urban residential communities through a neural network model, and belongs to the technical field of power systems. According to the technical scheme, the method comprises the following steps that (1) samples are collected; (2) demand factor samples are constructed; (3) the neural network model is trained; (4) the demand factors are calculated through the neural network model, it is assumed that the household number scale of a certain electricity utilization group is Q, the average house type area is A, and the occupancy rate is lambda, the household number scale, the average house type area and the occupancy rate are adopted as the input of the BP neural network model, and the household number demand factor Z(N) and the area demand factor Z(A) of the electricity utilization group are adopted as the output of the model. The method has the advantages that the community electricity utilization demand factors under the factor influence of the community type, the household number, the house type area, the occupancy rate and the like can be calculated precisely under the condition of a certain electricity load index, the defects that the demand factor reference range given by existing guide rules is too wide and not accurate enough are overcome, and therefore distribution network planning is effectively guided, and economical and safe running of a distribution network is guaranteed.