Abstract: With the development of science and technology, knowledge flow is constantly occurring among multidisciplinary fields, leading to the frequent convergence of different technologies. Early identification and detection of it will help companies find opportunities promptly and adjust technological innovation strategies. The existing measurement and prediction research of technology convergence based on patent have not formed a complete system yet. Aiming at this gap, a framework of measuring technology convergence is proposed. Based on patent data, the framework analyzes it from the perspectives of polymerizability, coupling, and equilibrium. The overall network formed by the technology convergence and the technical subjects that constitute the network are analyzed in detail. Finally, the equilibrium degree of the network is also studied. The model is verified by the convergence of pure electric vehicle technology and information technology in the Derwent patent database. Finally, based on the three dimensions, it is possible to screen out technologies that have a high level and a wide range of convergence, as well as a relatively stable technology portfolio. Relevant companies and R&D structures can conduct selective research and development according to the selected optimal technology or portfolio and promote the convergence and upgrading of related industries.
Nowadays, with the development of the digital economy, the complementary effects between technologies are constantly increasing, and the technological intersection between different industries is becoming more and more obvious. A single technology can no longer dominate a field, and single-subject knowledge can no longer meet the rapid development of society. At the same time, a single organization will meet with many uncertainties in the face of complex and changeable environment. Technology convergence drives technological improvement and industrial upgrading through mechanisms such as infiltration, crossover, and reorganization. It can also promote the interaction between different disciplines and provide a basis for solving various technological problems. Technology convergence also promotes cooperation among different organizations and improves innovation efficiency.
However, the existing patent-based measurement of technology convergence and prediction research has not yet formed a complete system. In terms of objectivity, patent is an indicator of technological innovation, which has been widely used to measure innovative output and is a relatively accurate index. Therefore, this paper takes patent information as an effective data source based on the classification and shows the current situation of technology convergence in the period of 2004-2014 by using technology co-occurrence map in macro-measurement. In the micro-analysis, technology convergence is measured bypolymerizability, coupling and equilibrium. Polymerizability mainly focuses on the network composed of various technologies. Coupling mainly measures technology convergence by taking each technology as the main body. Finally, the equilibrium of the network formed by technology convergence is also measured. Among them, the polymerizability is measured from three aspects: network density, network centrality and Φ index. Compared to the polymerizability, the coupling is concerned with the correlation between a module and other modules, mainly from the perspective of intensity and coverage. The measure of equilibrium is based on the information entropy.
With the aggravation of environmental pollution and the shortage of resources, many countries begin to develop new energy industry vigorously. As an important project of new energy vehicles, pure electric vehicles have been supported by the government and received widely concerned by people in recent years, but their technical system is still immature. Therefore, this paper mainly explores the convergence status of information technology and pure electric vehicle technology and points out the existing research and development priorities, which can provide guidance for future research and development activities according to the convergence situation.
In the case study, patents of pure electric vehicle technology and information technology are collected between 2004 and 2014. In the technology co-occurrence map of macro-analysis, the trend of technology convergence during 2004-2014 is analyzed and divided into three periods. The combinations of technologies with strongest relationships are G06F-H02J, H03K-H02M and H04N-B60R in 2004-2007, and G06F-H02J, H04N-B60R and G08C-G01R in 2008-2011. Compared with the previous period, more fields of information technology and pure electric vehicles are merging, although the intensity is lower, but the types are increasing, which indicates that the technology convergence in this period is developing towards diversification. The most powerful ones in 2012-2014 are G06F-H02J, H04N-B60R and H04B-H02J and the types of convergence in these two fields are also increasing compared with the period of 2008-2011. In microscopic analysis, from the three dimensions ofpolymerizability, coupling and equilibrium, it can be seen that G05F-H02M, G05F-H02J, G06F-B60L and G07C-B60W have higher convergence degree and quality, and they also have better development prospects. As for single technology, the distribution rate and coverage of B60W, B60R and G06F are at a high level, which indicates that there are many kinds of technologies converging with them, and the number of technologies they converge with is also increasing. From the perspective of equilibrium, in recent years, the convergence of information technology and pure electric vehicle technology is in an undulate and variable stage, and the most influential are H02J and G06G. In future R&D, relevant researchers and R&D institutes can select technology portfolios in strong convergence situation and emerging areas according to quadrant maps, which can guarantee a high rate of return. It is also possible to work on technologies with strong stability, so that the relevant investment can obtain relatively stable benefits.
The measurement of future technology convergence can also be further improved. In addition to the patent co-occurrence, it can also take the cooperation between institutions and individuals in different fields into account to further explore technology convergence. At the same time, it is also possible to consider the scenario where more than two technologies are converged, so as to more comprehensively study the future convergence trend.
In summary, this paper gives a new definition of technology convergence through patent-based measurement method, which lays a foundation for future patent-based measurement. Secondly, this paper proposes a relatively complete measurement method of technology convergence, which not only considers the overall internal convergence of the network formed by technology convergence, but also considers the different technologies that constitute the network and the coupling between different modules. Finally, the equilibrium of the network is explored.

Key words: technology convergence, polymerizability, coupling, equilibrium