Research On Identification Of Potential Disruptive Technology Driven By Multi-source Data

Technological innovation is the first driving force of economic development,and disruptive technology,as an important content of technological innovation,is the special kind of technologies that breaks the original technology life cycle,builds a new technology track,and replaces the existing mainstream technology in an unexpected way.Disruptive technology can realize the leap of social and technological system,change the performance of technical products,update the original technical performance standards,and have transformative significance for the military,science and technology,and industry.Countries around the world generally attach importance to disruptive technology research and development,and design special organizations or research and development plans to promote the development of disruptive technologies.However,disruptive technology development is highly uncertain,the research and development process is long.Research on identifying disruptive technologies early and accurately in a complex technical environment is of great significance to accelerate disruptive technology research and its development and grasp the development opportunities of international competition.This paper mainly analyzes and researches the identification of potential disruptive technologies,firstly sorts out the domestic and foreign research results of disruptive technology identification,and briefly expounds the basic concepts,theories and research methods involved in the research work of this paper.Secondly,after in-depth analysis of the connotation and characteristics of disruptive technology,combined with the feature indicators involved in the existing research results and the opinions of expert interviews,a characteristics indicator system of potential disruptive technology containing 5 first-level indicators and 17 second-level indicators is constructed,and the calculation method of each characteristic indicator is explained in detail.Then,taking the traditional BP neural network as the basic evaluation model,the whale optimization algorithm is introduced to improve it,and a CIWOA-BPNN identification model for potential disruptive technologies is proposed.Finally,taking the field of intelligence manufacturing as an example,relevant multi-source data are collected for technology theme analysis and potential disruptive technology identification,and then empirical analysis is completed.The results show that the potential disruptive technology identification model based on CIWOA-BPNN constructed and proposed in this paper has certain advantages in robustness,generalization ability and convergence speed.In addition,according to the results of disruptive technology identification,this paper analyzes and discusses the development trend,technology system and type of disruptive technology in the intelligence manufacturing.The characteristics indicator system of potential disruptive technology constructed in this paper mainly consists of computability indicators,which can provide a good theoretical reference for data-driven quantitative technology identification methods.At the same time,this paper expands the application of neural networks in the identification of disruptive technologies,providing new ideas for the research work of using non artificial methods to identify disruptive technologies.Finally,this paper conducts empirical research on the identification of potential disruptive technologies in the field of intelligent manufacturing.The research results not only help to more accurately grasp the development opportunities of the era in the field of intelligent manufacturing,lay out technology strategies in advance,but also provide a certain theoretical reference for the identification of disruptive technologies in other fields.