Structural changes on the complex network and the resulting knowledge and information flow varia_x005ftion are hot issues of current study．To analyze the accumulation process of the different knowledge on the heterogeneous complex network，the corresponding network model as well knowledge accumulation model is formulated in this paper．The change strategies of the heterogeneous knowledge network are divided into two categories． The first is equilibrium strategy，according to which the average account and the strength of connecting links between agents on the knowledge network remain fixed during the network evolution process．The second is the network growth strategy，which refers to that the connecting links between agents of the knowledge network are gradually increased．Correspondingly，the cumulative amount of all kinds of knowledge is categorized into two parts．The first part is the self-learning knowledge that is characterized by an S-shaped growth curve．The second is the external knowledge calculated according to the correlation matrix and the neighbor relationships． Simulation results show that a higher correlation degree of knowledge gets a faster flow speed in the network’s growth strategy and there is no evidence in the network’s balance strategy．Moreover，the variance of knowledge in the network’s growth strategy is smaller than that in the network＇s balance strategy