The invention discloses a method for predicting the power consumption growth rate through the economic growth rate, and relates to the technical field of power load prediction. According to the method, correlation inspection is carried out according to data between industrial development and electricity consumption of each department, and industrial development data related to the electricity consumption is obtained. Firstly, according to data between industry development and electricity consumption of each department, correlation inspection is carried out on economic indexes and electric power indexes, and correlation between the economic indexes and the electric power indexes is determined; and carrying out logarithmic processing on the data according to the annual growth rate and the daily power consumption of each department in the monthly industry. And according to the self-coding network, constructing a nonlinear regression model between economic variables to obtain an economic prediction power model based on nonlinear regression. The accuracy of the economic prediction power model is higher than the result of the logarithmic linear regression model. According to the technical scheme, the nonlinear regression model which is higher in precision and predicts the power consumption growth rate through the economic growth rate is adopted, and the prediction precision is high.