Abstract: 【Purpose/significance】The research aims to explore the relationship between the topological structure and the knowledge at⁃tribute of citation network and the effect of knowledge flow in citation network from the perspective of citation content.【Method/pro⁃cess】Firstly, on the basis of characteristics of knowledge flow in citation network from the perspective of citation content, the evolu⁃tionary dynamic model of knowledge flow is constructed from two aspects: static characteristic attribute and dynamic structure attri⁃bute; Secondly, according to main factors influencing the effect of knowledge flow in citation network, the mathematical model of the ef⁃fect coefficient of knowledge flow is constructed from three aspects: knowledge absorption, knowledge superposition and knowledge cost; Finally, the method to calculate the effect coefficient of knowledge flow is proposed by using the beehive model and its modified
model to describe knowledge spillover effect【. Result/conclusion】Netlogo is used to set the initial parameters of the evolutionary dy⁃namic model, and the influence mechanism of each factor on the effect coefficient of knowledge flow is analyzed through simulation ex⁃periments.The method in this paper can provide an empirical model for objectively reflecting the law and pattern of knowledge flow in citation network and revealing the effect of knowledge flow in citation network【. Innovation/limitation】This paper systematically analyz⁃es the knowledge flow effect of the citation network from the perspective of citation content through computational experiment methods, and draws a series of conclusions, but it does not compare and optimize the calculation results of actual bibliographic data.