The invention relates to a monthly electricity consumption prediction method of a seasonal support vector regression model, which comprises the following steps: obtaining historical electricity chargebill data of a user, converting the bill data, and converting the bill electricity quantity into monthly electricity quantity; Dividing a data set according to the data integrity degree of the contract number and the monthly electric quantity, and distinguishing effective users and invalid users; Processing the missing value and the abnormal value of the effective user data set; Calculating the seasonal index of each data sample, and correcting the original data by using the seasonal index; Constructing a support vector SVM time sequence feature matrix; And establishing an SVM for predictionanalysis and correction to obtain a predicted value. By implementing the method, monthly electricity consumption prediction of large industrial and commercial users is acquired, so that the users arehelped to know own power consumption conditions and realize node source open flow, and the method and the system are of great significance to electric power departments and are mainly beneficial to resource allocation optimization of the electric power departments and market development of electric power marketing departments.