In order to solve the problem of unbalanced electricity supply and demand of urban villages, the historical load data of an urban village in Guangzhou（N village）from 2016 to 2017 are analyzed. According to the relevant indicators of daily, monthly and quarterly（yearly）load characteristics, the change law and characteristics of the load of N village are determined. The load influencing factors of the urban village are classified, and SPSS is used to analyze the factors, conduct principal component analysis and complete linear regression, and the main influencing factors and load forecasting models that affect the load changes are obtained. The interpolation extrapolation method is used to verify the versatility of the model. The prediction of the load can provide the decision-making basis for the power supply department.