The use of symbol attributes on the side of symbolic social networks to analyze, understand, and predict the topology, function, and dynamic behaviour of complex networks, and has important theoretical significance for personalized recommendations, attitude prediction, user feature analysis, and clustering and application value. However, due to the huge scale of online social networks, this poses a challenge to traditional symbolic social network analysis methods. Based on the theory of structural equilibrium, this paper studies the evolutionary dynamics of symbolic social networks, proposes the energy function of weak structural equilibrium theory, and uses the evolution of evolutionary algorithms to obtain the weak imbalance of the network. The simulation experiment results show that the calculation method in this paper can get the optimal solution faster. It provides an idea for the study of real and complex social networks.