Abstract: The existing collaborative innovation network research mainly focuses on the construction of a single homogeneous collaborative innovation network,and most of them have not fully considered the interaction between multiple different subjects and their interactive evolution.Therefore,by decoupling a single homogeneous network with the super network theory and considering the problem of subject heterogeneity and the factors influencing network evolution,a multi-subject collaborative innovation evolution model of cross-level network containing cooperation network,knowledge network and innovation outcome layer was constructed,and a simulation analysis of cross-level network evolution and innovation performance change under different cooperation subject selection strategies were conducted.It was found that the new round of knowledge transfer of the subject would increase the connection of knowledge nodes,and the distribution of the connection between the cooperative network and the knowledge network was complex and increasing.The network had a high average knowledge level in the preferred connection mechanism based on node degree and supplemented by knowledge distance,which could promote innovation performance.Finally,the management strategies for partner selection,encouragement of core enterprises and universities,scientific research institutions and reasonable control of knowledge distance were presented.

Key words: collaborative innovation, cross-level network, network evolution, management strategy

摘要： 鉴于已有协同创新网络研究主要集中于构建单一同质协同创新网络,且大多研究未充分考虑多个不同主体之间的相互影响及其交互演化,结合超网络理论,将单一同质网络进行解耦,并考虑主体异质性问题和网络演化影响因素,构建了包括合作网络、知识网络和创新成果层的跨层次网络的多主体协同创新演化模型,并在不同合作主体选择策略下进行了跨层次网络演化、创新绩效变化的仿真分析。研究表明,主体新一轮知识转移会增加知识节点连边,合作网络与知识网络连线呈复杂、增长的趋势;网络在以节点度为主、知识距离为辅的择优连接机制具有较高的平均知识水平,能够促进创新绩效。最后给出选择合作伙伴、鼓励核心企业和高校/科研机构以及合理控制知识距离的管理策略。

关键词: 协同创新, 跨层次网络, 网络演化, 管理策略