The aggregate load online modeling with daily load profile has two key points to be solved， one is the appropriate classification and effective synthesis of customers' daily load profiles， the other is the industry composition ratio of the daily load profile in substations. Applying the theory of fuzzy C means clustering and pattern recognition， the authors propose a new method for consumer industry classification and synthesis with daily load profile of the power consumer. First， the customers are classified and selected on the basis of the active power in order to obtain the synthetical daily load curve of the industry. Then， the eigen-value parameters are obtained， by which the synthetical characteristics of the industry can be described and the results of classification and synthesis can be tested. Actual application shows that the method is clear in concept， convenient and practical.