Under the background of harmonious coexistence between man and nature, green innovation, as an important engine for green transformation of tourism industry, is a positive response to the improvement of ecological environment problems and the transformation of tourism development mode, and is directly related to the coordinated development of tourism economic growth and resource environment. The frequent flow of factors promotes the enhancement of green innovation links between cities. Exploring the path of matching the advantages of regional tourism green innovation network is helpful to promote the sustainable development of regional tourism. This paper constructs an analytical framework for regional tourism green innovation development, comprehensively applies the Super-SBM model, social network analysis, Haken model and kernel density estimation. Then chooses Shandong Peninsula urban agglomeration as the case study to measure and analyze its tourism green innovation efficiency, network dominance and their synergistic characteristics. Further, asymmetric innovation theory and qualitative comparative analysis are used to identify the configuration path for matching regional tourism green innovation development network dominance. The study found that: the green innovation efficiency of the tourism industry has the agglomeration distribution law of "coastal border" in space; there is a significant"Matthew effect"in the network dominance of tourism green innovation efficiency, forming a dotted pattern dominated by Jinan and Dongying; there are obvious regional differences in synergistic degree between the two, which can be classified into three categories, such as high synergy, moderate synergy and low synergy;the green innovation efficiency, network dominance and the matching promotion path of tourism industry can be summarized as seven modes, which is the outward market expansion mode supported by the government, the government-supported endogenous power-driven mode, and the dynamic optimization and adjustment mode under the dual-driven internal and external environments, among other things.