Identification of technology innovation path based on multi-feature vector fusion: The case of flywheel energy storage technology

Ziye Zhang, Lijie Feng, Jinfeng Wang, Weiyu Zhao and Jingbo Yan

Technological Forecasting and Social Change, 2025, vol. 212, issue C

Abstract: Flywheel energy storage (FES) technology, as one of the most promising energy storage technologies, has rapidly developed. It is essential to analyze the evolution path of advanced technology in this field and to predict its development trend and direction. However, some limitations remain in the existing research, which only uses a single feature to analyze technological innovation, fails to consider the development characteristics of technological innovation, and disregards the whole process analysis of the development trend of FES technology and the prediction of future development trends. Therefore, this study proposes a framework for technology evolution path identification and analysis that uses multisource data and incorporates citation and text features to monitor the evolution trend of FES technology and predict the future development direction of this technology. First, text and citation feature vectors from multisource data are extracted using shallow neural network embedding technology and then fused and spliced to obtain high-dimensional vectors that represent documents. Second, the time series of academic papers and patents filed in the last two decades are divided by the change point detection algorithm. Third, the Latent Dirichlet Allocation (LDA) model is applied to identify the topics of academic papers and patent data in different periods, and the cosine similarity calculation method is employed to construct the technical evolution path based on academic papers and patent data. Last, the gap between science and technology is analyzed, and the future development direction of FES technology is clarified.

Keywords: Flywheel energy storage (FES) technology; Shallow neural network; Latent Dirichlet Allocation (LDA) topic model; Change point detection (CPD) (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0040162524007649
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:tefoso:v:212:y:2025:i:c:s0040162524007649

DOI: 10.1016/j.techfore.2024.123966

Access Statistics for this article

Technological Forecasting and Social Change is currently edited by Fred Phillips

More articles in Technological Forecasting and Social Change from Elsevier
Bibliographic data for series maintained by Catherine Liu ().