基于应用场景的未来技术识别

Future Technology Identification Based on Application Scenarios

[研究目的]未来技术作为传统产业的转型动力及未来产业的形成基础,从技术的应用及其产业化视角出发,进行未来技术的识别与预测,对我国布局未来产业、增强发展优势具有重要意义。[研究方法]从技术单元与技术方案出发,构建基于应用场景的未来技术识别模型。首先,利用LDA主题模型对专利的用途进行主题聚类,识别技术的主要应用场景;其次,利用KeyBERT算法从专利的标题和新颖性文本中提取技术方案关键词,从技术方案的新颖性、关联性和重要性出发,筛选前沿技术方案;最后,定义产生前沿技术方案的技术单元为潜在未来技术,构建“技术影响力-技术生长力”坐标图,进行未来技术识别。[研究结论]以固体氧化物燃料电池领域为例进行实证研究,识别出13项未来技术,主要为SOFC批量化制造、可控性运行和实际化应用的技术,揭示了识别方法的有效性。

[Research purpose]Future technologies serve as the catalysts for the transformation of traditional industries and the building blocks for the future industries.It is of great significance to identify and predict future technologies from the perspective of the application of technology and its industrialization,because they play a pivotal role in shaping our nation's strategy for the future industries and enhancing our competitive advantages in development.[Research method]This paper constructs an application scenario-based model for identifying future technologies in terms of technology units and technological solutions.Firstly,we identify the main application scenarios of the technology by employing Latent Dirichlet Allocation(LDA)to conduct topic clustering of the patents based on their intended uses.Secondly,we employ the KeyBERT algorithm to extract keywords from patent titles and novel text,enabling us to pinpoint frontier technology solutions based on their novelty,relevance and importance.Finally,we define the technology units that generate frontier technology solutions as potential future technologies,construct a"technology influence-technology growth"coordinate diagram,and identify future technologies.[Research conclusion]Taking the field of Solid Oxide Fuel Cell(SOFC)as the empirical field,the effectiveness of the identification method is validated,through which we successfully identified 13 future technologies,primarily centered around SOFC batch manufacturing,controllable operation and practical application.