Leveraging heterogeneous networks to analyze energy storage systems in power systems and renewable energy research: a scientometric study

Bowen Dong, Zhen Guo, Aliya Mulat, Yuanhao Tian, Mingyu Lu, Yixuan Yuan, Xianchen LiuFrontiers in Energy Research 2024 · DOI

TL;DR

This study provides a comprehensive "map of the forest" for Energy Storage Systems (ESS) and renewable energy research over the last decade (2014–2023). Using advanced bibliometric tools, the authors reveal a massive surge in publication volume led by China and identify a critical shift in research focus: moving from basic grid integration toward system stability, efficiency, and AI-driven management.

What Problem Does This Paper Address?

The global transition to renewable energy is non-negotiable for sustainability, but it comes with a massive technical headache: intermittency. Energy Storage Systems (ESS) are the "glue" that makes renewable grids reliable, yet the sheer volume of research in this field has become overwhelming.

The authors argue that while individual breakthroughs are well-documented, the field lacked a high-level "scientometric" analysis—a study of the research itself. Without this, it is difficult for policymakers and engineers to identify which technologies are maturing, which countries are leading the charge, and where the next "research frontiers" lie. This paper fills that gap by analyzing the interconnectedness of thousands of data points to visualize the evolution of the discipline.

Methods

The researchers employed scientometrics, a subset of bibliometrics that uses mathematical models to analyze scientific literature. The primary tool used was CiteSpace (v. 6.1.R6), which allows for the visualization of complex academic networks.

Key techniques included:

  • Burst Detection: An algorithm used to identify keywords or citations that experience a sudden surge in frequency, signaling a "hot topic" or a shift in the research frontier.
  • Betweenness Centrality: A metric used to identify "bridge" nodes (countries, institutions, or papers) that connect different clusters of research. High centrality indicates a major influence on the field's direction.
  • Heterogeneous Network Analysis: Mapping the relationships between disparate entities, such as how specific institutions in one country drive specific keyword trends globally.
  • Timeline Visualization: Organizing keywords by year to show the "life cycle" of research themes.

Datasets

The study relied on a curated dataset from the Web of Science (WoS) Core Collection (specifically SCIE and SSCI).

  • Search Parameters: The authors used the string TS = “power system*” AND “new energy*”.
  • Filtering: After removing irrelevant entries (editorials, letters, etc.) and non-English articles, they arrived at a final pool of 425 high-quality research articles published between 2014 and 2023.

Key Insights and Ideas

  • The Geopolitical Gravity of Research: China has established a dominant position in this field, not just in volume (344 articles) but in influence. Its "centrality" score of 0.69 far outstrips the US (0.53), suggesting that Chinese research acts as the primary hub connecting various global sub-fields.
  • From Integration to Optimization: The "Aha!" moment in the timeline analysis shows a clear evolution. Between 2014 and 2018, the community was obsessed with "Integration" and "Management"—essentially figuring out how to plug batteries into the grid. Since 2019, the focus has pivoted sharply toward "Efficiency," "Stability," and "Photovoltaic Systems."
  • The Rise of the "Virtual" Power Plant: Recent "burst" keywords like Virtual Synchronous Generators and Reliability Assessment indicate that the field is moving away from purely hardware-based solutions toward sophisticated software and control strategies to mimic traditional power plant behavior.
  • Institutional Heavyweights: A few key players drive the majority of the innovation. North China Electric Power University, Tsinghua University, and the China Electric Power Research Institute emerged as the most densely connected nodes in the institutional network.

Key Results

  • Exponential Growth: The field grew from a mere 3 related papers in 2014 to 125 in 2023.
  • Top Categories: Engineering (37.3%) and Energy Fuels (19.9%) dominate the landscape, though Computer Science (5.03%) is a growing slice of the pie.
  • Most Influential Work: The most cited paper in the dataset is Quilumba et al. (2015), which focused on using smart meter data to improve load forecasting—a foundational requirement for efficient ESS deployment.
  • Collaboration Gaps: While China and the US collaborate extensively, the study found that countries like Spain, Poland, and Denmark have very low centrality, suggesting they are conducting "siloed" research that isn't yet well-integrated into the global network.

Limitations and Future Work

The authors acknowledge several constraints:

  • Database Bias: By relying solely on Web of Science, the study may have missed significant contributions published in Scopus, IEEE Xplore, or non-English language journals.
  • Retrospective Lag: Bibliometric analysis is inherently backward-looking. It captures what has happened, potentially missing "bleeding-edge" developments from the last six months.
  • Future Directions: The paper suggests that future research must prioritize:
  • AI/ML Integration: Using deep learning for real-time ESS demand/supply forecasting.
  • Advanced Chemistry: Moving beyond standard lithium-ion to solid-state and flow batteries for better safety and lifespan.
  • Hybrid Solutions: Exploring the synergy between different storage types (e.g., combining thermal and electrical storage).

Summary updated

Abstract

The transition to renewable energy sources is critical for sustainable development, yet integrating these sources into existing power systems poses significant challenges. Energy Storage Systems (ESS) are essential in enhancing the reliability and efficiency of renewable energy systems. Despite growing research, a comprehensive scientometric analysis mapping development and trends in this field is lacking. This study addresses this gap by conducting a detailed scientometric analysis of power systems and new energy research from 2014 to 2023. The novelty of this study lies in its systematic use of advanced bibliometric tools to provide a thorough analysis of the research landscape. Utilizing 425 research articles from the Web of Science database, the study employs CiteSpace to visualize academic networks, identify research hotspots, and outline current trends. Specific methodologies include burst detection to identify significant shifts in research focus, centrality measurement to determine the influence of key studies, and heterogeneous network analysis to map the interconnectedness of various research themes. The analysis reveals extensive international collaborations, with China leading in publication volume (344 articles) and centrality (0.69), followed by the United States (29 articles, centrality 0.53). Significant contributions come from institutions like North China Electric Power University, China Electric Power Research Institute, and Tsinghua University. The findings underscore the importance of international cooperation and the need for broader geographical representation in this research field. This study provides valuable insights into the evolution and current state of power systems and renewable energy research, offering essential guidance for future research and development. The results highlight the critical role of ESS in the transition to renewable energy and suggest directions for future investigations, particularly focusing on enhancing ESS efficiency and reliability and expanding international research collaborations.
BibTeX
@inproceedings{dong2024leveraging,
  author = {Bowen Dong and Zhen Guo and Aliya Mulat and Yuanhao Tian and Mingyu Lu and Yixuan Yuan and Xianchen Liu},
  title = {{Leveraging heterogeneous networks to analyze energy storage systems in power systems and renewable energy research: a scientometric study}},
  booktitle = {Frontiers in Energy Research},
  year = {2024},
  doi = {10.3389/fenrg.2024.1424928},
}
Figure