Abstract: The existing researches on innovation search are mostly built on the un-weighted static network, and assume that the innovation search and knowledge transfer are two independent processes. In order to reproduce the overall behavior mechanism of innovative search, knowledge transfer and knowledge innovation more realistically, this paper adopts the method of simulation modeling based on the scale-free weighted dynamic network, to reveal the evolution of the dynamic growth of knowledge, explore the intrinsic relationship between innovative search, innovation network and innovation performance. The study found: No matter what kind of search strategy, the innovation network evolution is basically the same, and the final network evolution will reach equilibrium; there is an inverted U-shaped relationship between the network connection density and average network knowledge stock, and network tie strength and network average knowledge stock is positively correlated, while the innovation search strategy has a significant impact on the average knowledge stock and the level of knowledge innovation.