The Formation,evolution,and Effects Of R&D Collaboration Network Toward Low-carbon Energy Technology

Energy is a paramount strategic requirement pertaining to the sustainable development of national economy in the contemporary society.The development of low-carbon energy technology has developed into the key path to solve the worldwide energy shortage issue and consequently has gained high degree of concern and support in every country.Low-carbon energy technology is a complex systemic engineering which manifests in its multidisciplinary,multi-domain and sector-spanning features.These features are accounted for its dependence on the R&D collaboration network as it provided a mainstream carrier as well as platform in the technology development,diffusion and application efficiency evaluation.However,the existent literature is scattered and only focus in single industry or partial achievement.The systematic and multi-factor controlling underlying rules of R&D collaboration network toward low-carbon technology are remained to be explored.Therefore,this research applies green patent codes disclosed from IPC Green Inventory of The World Intellectual Property Organization(WIPO)as the objective data source,and conducts a system research of low-carbon energy technology R&D collaboration network(hereafter referred to as LCET-CN)in the perspectives of its formation,evolution,structural effects,and optimization of enterprise network strategies as well as the policy orientation.The research outcomes and innovation points are as follows,(1)Build the formation mechanism model of LCET-CN and conduct simulation to shed light on its emergence propertiesRegarding the complex adaptive system property of LCET-CN,build the multi-agent model based on complex network.This is complemented with simulation to analyze system optimization process to shed light on the motivation mechanism and its emergency property along time of the R&D collaboration network.This model turns the network linking mechanism into a dynamic optimization model,and applies Matlab to perform the Differential Evolution Algorithm to solve the optimization model and further to execute the simulation.The simulation data obtained are analyzed statistically to reveal complex network characteristics and organization-level emergence properties.Results shows that differences exists between LCET-CN and R&D collaboration network toward traditional technology in terms of scale-free property and the degree as well as direction of network structural effects,thus indicating the comparative significance of social network structure effects in LCET-CN analysis.This not only overcomes the strict hypothesis about static network in the complex network based evolutionary game methodology in the existing studies,but also offers theoretical foundation for the planning and building of highly efficient eco-environment for R&D collaboration network.In the meantime,based on the formulation of R&D investment rule,knowledge absorption rule,knowledge formation rule,and innovation as well as profit generation rule,this model makes the exogenous partnership selection in the previous studies endogenous,and overcomes the vague interpretation of node linking principle in the current literature.(2)Construct the LCET-CN characterized by joint patent filing and reveal its evolution rulesIdentify the technology range of LCET-CN based on 212 patent codes belonging to 9 low-carbon categories disclosed from IPC Green Inventory of The World Intellectual Property Organization,use Java language to exploit joint patent filing data and generate adjacency matrices,and then construct the LCET-CN comprised by 491 organizations in mainland China during 2006 to 2015.Furthermore,analyze the evolution characteristics in terms of global network property,actors' positions,and collaboration relationship.The results indicate that global network exhibits the properties of scale-free network and small world maximal connected subgroup;the identification of central actors as well as their evolution characeritics implies that the government should implement stratified stimulation policies to central actors belonging to 9 catogories of LCET-CN in the heterogeneos industrial development stages;the collaboration relationships analysis has on the one hand indicated that relationship intensity of industrial-academic-research cooperation as well as sector-spanning collaboration should be increased,and has on the other hand exhibited larger technological distance in terms of collaboration relationships,indicating a higher standard required in absorptive capability of R&D collaboration parties.This is the first quantification analysis of LCET-CN based on joint patent filing data from the angle of innovation technology orientation rather than sector perspective.(3)Establish panel data count model to perform empirical confirmation of LCET-CN structural effectsApply the embeddedness theory and so on to analyze the individual as well as the interaction effects of global network property,actor position and collaboration relationship on the exploratory and exploitative innovation,and propose the research hypotheses.Establish the panel data comprised of 757 firm-year observations of 215 enterprises and further apply panel data count model to perform empirical analysis.The results shows that cultivation of network clustering coefficient contributes more to technology innovation than the improvement of density;actor position decision should be matched with global network property and collaboration relationship;the direction and degree of interaction effects of LCET-CN structural properties differs between exploratory innovation and exploitative innovation;the improvement of collaboration intensity is conducive to exploratory innovation and exploitative innovation.This has not only fills the empirical gap regarding the effects of R&D collaboration network in the context of low-carbon technology innovation,but also exploits the specialties of LCET-CN structural effects in-depth in the perspective of the interaction effects of threefold structure on exploitative as well as exploratory innovation.Therefore,the general rules of advancement in LCET-CN structural effects could be illuminated based on this analysis.(4)Put forward the network strategy and policy orientation regarding to the improvement of LCET-CN structural effects based on coupling analysisCombine the three-dimensional plots and two-dimensional matrices to reveal the enterprise network strategies and government network governance orientation in the promotion of technology innovation in LCET-CN.Specifically,the three-dimensional plots are drawn out from the interaction effect between actor position and collaboration relationship,and global network property and actor position.These three-dimensional plots are further converted into the two-dimensional matrices to reflect the coupling relationship between interaction type and technology innovation level.And then two-dimensional matrices are used to infer enterprise network strategies and their realization paths,and government network governance orientation as well as related policies.On the one hand,this analysis offers operable references for the enterprises with heterogeneous strategic stances and resource capability to make suitable network strategies in line with its innovation appeals.On the other hand,this analysis provides decision support for the government in the scientific and sound application of policies in terms of hierarchy and phases.