基于Nooteboom模型构建了网络中企业间创新合作的博弈模型，分析了该博弈的风险占优均衡，确定企业合作伙伴数的临界值，并基于复杂网络模型仿真分析了企业网络的动态演化，研究企业网络结构特征与合作度的变化规律。研究结果表明：企业合作的临界伙伴数量由合作的收益、成本及合作风险决定。当企业合作伙伴数量小于临界值时，合作是企业的最优策略，并且随着合作度的增加，合作创新网络的特征路径长度缩短，集聚度下降趋于稳定值，合作创新网络出现均匀而致密的结构；在企业合作的伙伴数等于临界值时，合作创新网络中的创新合作度达到最大并趋于稳定；当企业合作伙伴数量大于临界值时，企业间的创新合作度将下降。

This paper proposed an innovation cooperation game model in the enterprises networks based on the modified Nooteboom model. We analyzed the risk dominant equilibrium of the game model and obtained the threshold of the partner number in the enterprises networks. Furthermore, we simulated the evolutionary dynamic of the enterprises innovation cooperation networks and learned the rules of their topology properties based on the theory of complex network. The results indicated that the thresholds of the partner number in the enterprises networks were depended on the benefits, costs and risks of innovation cooperation. When the numbers of enterprises’ partners is smaller than the threshold, the cooperation rates will increase and the characteristic path lengths will shorten and the cluster degrees will decrease. When the numbers of enterprises’ partners is equal to the threshold, the cooperation rates will reach the maximum and tend to stability. When the numbers of enterprises’ partners is larger than the threshold, the cooperation rate will decrease.

F425

国家自然科学基金项目《基于超网络的企业多主体合作创新与创新群体演化研究》（71373198）；湖北省科技支撑计划软科学项目《企业间合作创新资源激励及横向联盟网络建立的研究》（2014BDF103）；武汉体育学院青年教师科研基金项目（2015QS06）