Mapping the Evolution of New Energy Vehicle Fire Risk Research: A Comprehensive Bibliometric Analysis

Abstract

To gain a comprehensive understanding of the current research landscape in the field of new energy vehicle (NEV) fires and to explore its knowledge base and emerging trends, bibliometric methods—such as co-occurrence, clustering, and co-citation analyses—were employed to examine the relevant literature. A research knowledge framework was established, encompassing four primary themes: thermal management and performance optimization of power batteries, battery materials and their safety characteristics, thermal runaway (TR) and fire risk assessment, and fire prevention and control strategies. The key research frontiers in this domain could be classified into five categories: mechanisms and propagation of TR, development of high-safety battery materials and flame-retardant technologies, thermal management and thermal safety control, intelligent early warning and fault diagnosis, and fire suppression and firefighting techniques. The focus of research has gradually shifted from passive identification of causes and failure mechanisms to proactive approaches involving thermal control, predictive alerts, and integrated system-level fire safety solutions. As the field advances, increasing complexity and interdisciplinary integration have emerged as defining trends. Future research is expected to benefit from broader cross-disciplinary collaboration. These findings provide a valuable reference for researchers seeking a rapid overview of the evolving landscape of NEV fire-related studies.

1. Introduction

The development of new energy vehicles (NEVs) has emerged as a key strategic focus for the global automotive industry in recent years, driven by the need to reduce emissions, conserve energy, and protect the environment [1]. International automobile companies, groups, and alliances have announced their electrification strategies, pushing the NEV sector into a period of rapid growth. With advantages such as low emissions, low noise, high energy efficiency, and long service life, NEVs have attracted increasing attention worldwide, surpassing traditional internal combustion engine vehicles (ICEVs) in popularity [2]. Several countries have adopted aggressive NEV promotion plans; for example, the Netherlands has proposed to end production of fuel vehicles by 2030, the European Union has prioritized EV deployment for carbon reduction, and Germany aims for nationwide EV adoption by 2050. In China, the NEV market experienced exponential growth under the guidance of the State Council’s 2020 Energy Saving and New Energy Vehicle Industry Development Plan (2021–2035), making it the largest NEV market globally [3]. According to statistics from the China Association of Automobile Manufacturers, China’s NEV sales reached 9.495 million units in 2023, accounting for 31.6% of the market and representing a 37.9% year-on-year increase. This growth trend is expected to continue into 2025. However, alongside the rapid expansion of the NEV market, concerns regarding fire safety have become increasingly prominent. According to data released by the National Fire and Rescue Administration of the Ministry of Emergency Management of China, 640 NEV fire incidents were reported in China in the first quarter of 2022, corresponding to an estimated incidence of approximately 0.7179 per 10,000 NEVs. For comparison, the reported fire incidence rate for other road vehicles excluding NEVs and electric bicycles in China during the same period was about 0.4892 per 10,000 vehicles. It should be noted that these figures exhibit regional specificity and should not be generalized globally. It is important to note that this figure includes all types of NEV fire incidents, not exclusively those caused by battery system failures. These fires have raised public safety concerns, caused notable property losses, and posed serious risks to both passengers and drivers [4]. A comparative analysis between NEV fires and those involving traditional ICEVs reveals significant differences in ignition causes, fire behavior, and suppression challenges. ICEV fires are typically caused by factors such as fuel leakage, mechanical friction, overheating of engine components, or electrical short circuits. These fires tend to follow more predictable patterns and are generally managed with well-established firefighting strategies. In contrast, NEV fires are often triggered by thermal runaway (TR) in battery systems [5]. The fire can propagate internally within battery modules and may lead to the release of toxic gases, making the fire behavior more complex and difficult to predict, as well as considerably harder to extinguish [6]. That said, from an operational firefighter’s perspective, many core firefighting tactics are similar to those applied to ICEV fires, and water remains the primary extinguishing medium in most cases. The main operational differences are that battery fires often exhibit much longer durations, carry a higher risk of reignition, and pose practical challenges in delivering sufficient cooling water to the affected cells or modules. These characteristics—prolonged heat release, potential for thermal propagation within cells, risk of rekindling, and hazardous off-gassing—require adjusted tactical responses such as extended cooling and sustained monitoring for re-ignition, as well as consideration of ventilation and isolation measures. Consequently, research is advancing on supplementary solutions such as targeted suppression agents, fine water mist, and improved detection and early warning systems to strengthen standard firefighting tactics and lower post-extinguishment risks. These differences underscore the necessity of dedicated research into NEV fire safety [7]. As a result, the prevention and mitigation of NEV fire incidents has become a critical and urgent area of research.

Scholars around the world have conducted extensive investigations on NEV fires through numerical simulations, experimental research, and theoretical studies. The causes of such fires are diverse, with the power battery identified as a key component and the primary source of many fire incidents. Zou, Li, and Lu (2022) examined the TR and combustion behavior of pouch cells under varying heat fluxes, establishing a descriptive model of flame temperature distribution that informs thermal hazard evaluation [8]. Similarly, Zhao et al. (2024) explored TR-induced fire spread in full-scale EV tests and assessed suppression strategies, including fire blankets, water sprays, and foams, providing practical insights for fire response [9]. Beyond fire behavior, preventive and suppression technologies have been a central focus. In the nitrogen–phosphorus flame suppression system, Zhan et al. (2024) added SiO₂ aerogel fragments as an operational filler [10]. The results showed that excessive amounts of SiO₂ aerogel decreased the performance of fireproofing coatings for EV battery packs. Hu et al. (2024) elucidated the ability of fine water mist in TR suppression in lithium-ion batteries (LIBs) [11]. Since fire experiments still had problems such as high cost, high safety risk, and poor repeatability, some researchers adopted predictable, operational, and reusable computer simulation techniques to simulate specific scenarios of fires in NEVs. Cheng et al. (2023) developed a coupled multi-area model to predict LIB thermal reactions under fire scenarios [12]. Bai et al. (2024) compared the heat release rate (HRR), calculating models to analyze combustion in NEVs and conventional vehicles within tunnels [13]. Chen et al. (2024) established a native finite element simulation (FEM) of a vehicle powered by electricity with the required components, proposing structural design improvements for battery fire safety [14]. In parallel, risk assessment has emerged as a critical research stream. Cho, Park, and Kim (2022) integrated electrochemical modeling with continuous monitoring to establish a real-time framework for evaluating fire risks in high-capacity battery packs [15]. To quantify the risk level of electrolyte leakage and other failure types, Zhang et al. (2023) analyzed electrolyte leakage mechanisms and applied AI-based multimodal fault diagnosis, enabling early detection of failures up to 26 days in advance [16]. Collectively, these studies demonstrate a transition from understanding TR mechanisms to developing predictive tools, fire prevention measures, and real-time risk management systems for NEV safety.

The above-mentioned studies have shown that substantial research has been conducted on specific applications related to NEV fires associated with LIB systems. Bibliometrics is a multidisciplinary approach that uses mathematical and statistical methods to analyze scientific publications and knowledge structures [17]. Bibliometric methods enable the analysis of the literature in terms of sources, co-authorship networks, keywords, citation relationships, and more. This approach allows researchers to efficiently grasp an overview of the field, including its knowledge base, research hotspots, and advancement patterns of that sector [18]. In this paper, bibliometric methodologies were adopted using tools such as CiteSpace [19], VOSviewer [20], and Bibliometrix R [21] to quantitatively analyze the research literature on NEV fires associated with LIB systems. This study systematically examined the temporal distribution of publications, as well as the contributions of key countries, journals, and research institutions. By analyzing the core literature, author contributions, high co-citations, and keyword co-occurrences, a comprehensive knowledge map was created, summarizing the research structure in this field [22]. Furthermore, the study identifies emerging research frontiers and outlines the evolutionary trajectory and development trends within the field. This work systematically integrates diverse perspectives to reveal hidden connections, developmental patterns, and collaboration structures. The findings not only provide a comprehensive understanding of the research landscape but also highlight theoretical underpinnings for battery fire prevention and practical implications for NEV safety management. It makes a distinctive contribution to bridging fragmented studies and guiding subsequent investigations into LIB-related fire safety.

In this study, NEVs are defined as vehicles powered mainly by LIBs, which are the dominant energy source for modern electric vehicles. According to the Ministry of Industry and Information Technology of China, NEVs include battery electric vehicles (BEVs), plug-in hybrid electric vehicles (PHEVs, including range-extended types), and fuel cell vehicles. Here, the term NEV fires broadly covers fire-related safety issues in BEVs, PHEVs, and other LIB-powered vehicles. This encompasses external fires caused by accidents or electrical faults as well as internal thermal hazards such as TR, thermal propagation, battery rupture, reignition, and explosions. Research on battery thermal safety, failure mechanisms, suppression strategies, and firefighting technologies is therefore considered within the scope of NEV fire studies. In this paper, the term ‘power battery’ refers to the propulsion battery used in NEVs.

To provide a clear roadmap of the manuscript, the structure is organized as follows: In Section 2, the materials and methods are described, beginning with Section 2.1, which outlines the data sources. Section 2.2 details the research methods and tools, including the bibliometric and visual analysis techniques employed. Section 3 presents the results and discussion, starting with Section 3.1, which explores the temporal distribution of the literature on NEV fires related to LIB systems. In Section 3.2, the spatial distribution of the literature is analyzed, with subsections covering the country and region distribution (Section 3.2.1), institutional distribution (Section 3.2.2), and major journal distribution (Section 3.2.3). Section 3.3 delves into the research knowledge base, with a focus on core author analysis (Section 3.3.1), high-cited literature analysis (Section 3.3.2), and analysis of highly co-cited journals (Section 3.3.3). Section 3.4 investigates research evolution and hotspots, including keyword co-occurrence and safety themes (Section 3.4.1), research frontiers in battery fire safety (Section 3.4.2), and the evolution of research topics and safety strategies (Section 3.4.3). Finally, Section 4 provides the conclusions, summarizing the main findings and suggesting directions for future research in the field of NEV fire safety.

2. Materials and Methods

2.1. Data Sources

The Web of Science (WoS) database is widely recognized as one of the most authoritative and comprehensive sources of English-language academic literature, with significant global influence [23]. To ensure data consistency and reproducibility and to provide a comprehensive view of the research landscape and development trends in NEV fires associated with LIB systems, this study selected the Science Citation Index Expanded (SCI-EXPANDED, 2008–2024) and the Social Sciences Citation Index (SSCI, 2011–2024) from the WoS Core Collection as data sources. The strict selection process of journals in WoS, together with its complete citation records and export functions, ensures its suitability for bibliometric mapping and co-citation analysis in software like VOSviewer 1.6.19. and CiteSpac 6.3.1. To avoid record duplication and metadata inconsistency when combining sources, we restricted the analysis to WoS.

Table 1. WoS database: search strategy table, 2008–2024.

Rank	Retrieval Strategies	Date Set	Number of Records	Periods	Dataset Used in Each Section
1	TS = (new energy vehicles)	A	11,919	2008–2024	Not used
2	TS = (lithium battery)	B	146,474	2008–2024	Not used
3	TS = (fire) OR TS = (fire disaster)	C	162,898	2008–2024	Not used
4	TS = (new energy vehicles) OR TS = (lithium battery)	A∪B	156,740	2008–2024	Not used
5	(TS = (fire) OR TS = (fire disaster)) AND (TS = (new energy vehicles) OR TS = (lithium battery))	(A∪B)∩C	1675	2008–2024	Section 3.1 Section 3.4
6	(TS = (fire) OR TS = (fire disaster)) AND (TS = (new energy vehicles) OR TS = (lithium battery))	(A∪B)∩C	1465	2019–2024	Section 3.2 Section 3.3 Section 3.4

A∪B: dataset contains all the articles in the A or B; A∩B: dataset with articles found in both datasets A and B; TS stands for ‘Topic Search’ in the Web of Science database.

lists six search strategies: the first four strategies were not used, and the search strategy (TS = (fire) OR TS = (fire disaster)) AND (TS = (new energy vehicles) OR TS = (lithium battery)) was analyzed comprehensively as the final search formula. To ensure a comprehensive and systematic retrieval process, six search strategies were initially formulated and assessed. The first four strategies were designed to explore broader or more generalized keyword combinations in order to evaluate their relevance and overlap with the core research theme. However, due to the excessive volume of literature retrieved, these strategies were excluded from the final analysis. Their inclusion in Table 1 serves to transparently document the refinement of the search process. Ultimately, the last two strategies, offering greater specificity in identifying the literature related to NEV fires associated with LIB systems’ incidents, were adopted for the bibliometric investigation. The search period ended on 31 December 2024, and 1465 documents from search method six were utilized as samples of information for research.

2.2. Research Methods and Tools

To perform quantitative analysis and visualization of the literature in the field of NEV fires associated with LIB systems research, this study adopted the bibliometric method to analyze a large body of research. Bibliometrics refers to a cross-disciplinary field that quantitatively analyzes all knowledge carriers using mathematics and statistics. In order to effectively identify the research trends of a particular field, as well as to evaluate and forecast the discipline’s future development, bibliometrics analyzes the quantitative features of the literature, examines the structural distribution of information, internal quantitative relationships, and the law of change.

Using software like VOSviewer, Bibliometrix R 4.4.1, and CiteSpace, the co-occurrence, cluster, co-citation, and keyword clustering analyses of the 1465 articles that were exported from the WoS Core Collection were examined in this paper using bibliometric and visual analysis techniques. VOSviewer was effective for building and visualizing citation and co-occurrence networks, which facilitated the identification of research hotspots and thematic groupings. CiteSpace was particularly useful for examining the temporal dynamics of the field, especially via co-citation networks and burst detection, allowing emerging topics and frontiers to be recognized. Bibliometrix R, an open source R package, offered a versatile platform for calculating bibliometric indicators, performing statistical assessments, and exploring the evolution of themes. The integration of these tools enabled a comprehensive examination of NEV fire research associated with LIBs from 2008 to 2024 and allowed its development path to be mapped. The specific research process is shown in Figure 1. In Section 3.1, the temporal distribution pattern of papers worldwide is analyzed. The geographical distribution of papers on NEV fires associated with LIB systems, including active nations and regions, significant research institutions, and major journals, is plotted in Section 3.2. In Section 3.3, the core authors, core literature, and highly co-cited journals in the field of NEV fires associated with LIB systems are visualized and analyzed. In order to determine the research hotspots and frontiers in the NEV fires associated with the LIB systems sector, the research frontiers and topics are outlined using the keyword co-occurrence analysis, which is carried out in Section 3.4.

3. Results and Discussion

3.1. Temporal Distribution of Literature

The number of published articles can reflect both future research trends and annual shifts in study topics. Figure 2 displays the statistical findings of the yearly grouping of the 1675 papers with the research topics of (TS = (fire) OR TS = (fire disaster)) AND (TS = (new energy vehicles) OR TS = (lithium battery)) that were obtained from the WoS Core Collection between 2008 and 2024. The number of annual publications, as well as the overall number of annual cumulative publications within the literature, is shown in Figure 2. The regression equations in Figure 2 describe the exponential growth trends of annual publications on NEV fires associated with LIB systems. The results confirm that the exponential trend effectively captures the historical development pattern of this research field. The overall trend of the number of publications on NEV fires from 2008 to 2024 demonstrated an upward trend. It can be observed from Figure 2 that the overall development process of research on NEV fires associated with LIB systems can be broken down into three stages: the initial exploratory stage (2008–2012), the stable development stage (2013–2018), and the rapid growth stage (2019–2024).

Initial exploratory stage (2008–2012): In this stage, the annual number of publications worldwide remained below 20, and the overall growth of the annual publication count was minimal. Research on fires in NEVs primarily focused on electrode materials and additives for power batteries. In 2008, Feng et al. (2008) investigated the novel electrolyte additive tris (4-methoxy thistle) phosphate, which had reduced electrolyte flammability and enhanced thermal stability, providing overcharge protection and flame-retardant effects for NEV power batteries with minimal impact on battery performance [24]. Walz et al. (2010) found that nanoporous Zirconia (ZrO₂) and Titania (TiO₂) coatings had stabilized the cycling performance of lithium manganate spinel cathode LIBs and had evaluated the final performance of the coated cathodes [25]. Additionally, solid polymer electrolytes with flame-retardant properties have attracted increased scientific attention, prompting further research [26].

Stable development stage (2013–2018): The number of annual publications continued to rise, with the number of articles issued from 2017 to 2019 growing at a faster pace. During this stage, research on TR and the fire behavior of power batteries for NEVs became more in-depth and extensive. Ping et al. (2015) discovered through experimentation that the battery’s charge level significantly impacted mass loss, total heat generation, and maximum HRR [27]. Chen, Yuen, and Wang (2017) experimentally investigated LIBs with different arrangements and analyzed the combustion behavior in greater detail by using mass-loss rate, HRR, and heat flux [28]. The study concluded that batteries with a larger exposed heating surface were more prone to intense fire events.

Rapid growth stage (2019–2024): In this stage, the annual publication volume had risen rapidly, reaching 403 articles per year in 2024, indicating that the field of NEV fires was expected to continue growing steadily with high research interest in the future. The degree of related research has become more in-depth. EV fires mainly originated from the TR, and the spread of fire triggered by the TR can lead to serious disasters. In contrast, fire blankets, water spray, and compressed air foam have certain suppression abilities for EV fires in the early stage, effectively reducing the temperature inside the vehicle and extinguishing the flames [9]. With the advancement of computer technology, machine learning has significantly broadened the scope and depth of research on NEV fires. For example, Biharta, Santosa, and Widagdo (2023) investigated a sandwich-based honeycomb structure using a machine learning approach [29]. This structure is designed to protect pouch cells in EV battery systems under axial impact loads. The method helps prevent severe deformation of LIBs during collisions, which could otherwise trigger TR.

3.2. Spatial Distribution of Literature

3.2.1. Country and Region Distribution

The spatial distribution of academic publications enables researchers to rapidly identify regions with strong research capacity, offering valuable insights for fostering scientific partnerships and facilitating the integration of research outcomes. Through the statistical analysis of literature sources by region or country, it becomes feasible to evaluate the research intensity and thematic influence in various geographical areas, which in turn promotes international academic cooperation and the sharing of expertise [30]. Furthermore, the spatial distribution of publications within a region reflects the extent of its present research emphasis on the topic. This study analyzed research articles on NEV fires associated with LIB systems from the WoS Core Collection, with the dataset encompassing academic publications from 60 different countries and jurisdictions.

Table 2. Top 10 countries with publications in the field of new energy vehicle fires, 2019–2024.

Rank	Country	Region	Quantity	Percentage	Total Link Strength	Total Citations	Average Citations
1	China	East Asia	777	51.87%	191	21,451	27.6075
2	USA	Northern America	197	13.15%	129	7170	36.3959
3	South Korea	East Asia	93	6.21%	45	1009	10.8495
4	UK	Western Europe	74	4.94%	67	2169	29.3108
5	Australia	Oceania	40	2.67%	46	1681	42.0250
6	Germany	Central Europe	33	2.20%	29	501	15.1818
7	India	South Asia	28	1.87%	28	721	25.7500
8	Japan	East Asia	27	1.80%	11	577	21.3704
9	Canada	Northern America	22	1.47%	29	788	35.8182
10	Spain	Southern Europe	22	1.47%	34	717	32.5909

displays a count of the top ten countries by the number of papers published. China (777), the United States (197), and South Korea (93) were the top three countries in terms of the number of published papers among the top ten countries. The total number of papers published by the three countries represented 71.23% of the total number of papers and over fifty percent of the overall number of published articles. In terms of the geographical distribution, the average number of references for the papers published in the US and Europe was typically greater. This suggested that the quality of research results was greater, the published papers were of a higher standard, and the field’s research was more important, well-known, and frequently cited. China ranked first in both the number of publications and total citations in NEV fire research, indicating a high level of research activity. However, the average citation count per article was only 27.6075, highlighting a gap compared to European and American countries that began related studies earlier. This discrepancy may be explained by several factors. First, journal visibility and accessibility play an important role: many highly cited journals in Western countries are open access or widely indexed internationally, increasing their exposure and citation likelihood. Second, language barriers may limit the readership of Chinese publications, as most highly cited papers in the field are in English, reducing their accessibility to a global audience. Third, differences in research maturity and historical development affect citation patterns. Western countries and North America initiated NEV fire and battery safety research earlier, leading to more established research networks and recognition. Finally, differences in research focus and methodological rigor may also influence citation counts, as papers that address broadly recognized technical challenges or employ internationally standardized methods tend to attract more citations. This suggests that the overall quality and academic influence of Chinese publications could be improved.

Among the top five nations in terms of the number of publications, two countries belonged to North America. Although Australia ranked only fifth in the number of publications, it exhibited a relatively high academic impact, with an average of 42.0250 citations per article. This suggests that research originating from Australia in this field has been comparatively well-developed and influential. Using VOSviewer software, the type of analysis was determined to be a cooperative relationship, and the field type was set to the country to generate a country/region cooperation network diagram, as shown in Figure 3. Each node in this diagram represents a different country, and the bigger the node, the more articles published; the width of the line that connects the nodes indicates the level of international cooperation. The node colors represent their respective clusters, with nodes of the same color indicating research directions that were closely related. It is evident that the United States and China contributed the most to the study of NEV fires. Furthermore, while having fewer publications than China and the US, European nations collaborated more closely within the regional distribution framework. The aggregate volume of articles from the top ten countries and areas accounted for 87.65% of the total number of articles published, according to an analysis of these high-producing nations and locations. At that time, the major research organizations in this sector were relatively unbalanced globally, with a handful of nations or areas producing most of the research results. These nations or areas were mostly made up of industrialized nations or major global economies. The main reason was that these countries had an early industrial start, NEVs had been developed earlier, and people had been concerned about NEV fires, so there was more research on NEV fires earlier than in other countries. The information obtained could be analyzed using VOSviewer and Scimago Graphica software 1.6.19 to create a country/region cooperation network diagram, as seen in Figure 4.

3.2.2. Institute Distribution of Literature

By analyzing institutional collaborations reflected in the literature, it is possible to understand patterns of academic cooperation, research directions, and the influence of research outcomes, thereby offering valuable insights to scientists working in related fields [31]. A total of 1203 institutions have conducted research in this area, based on data from the WoS core database.

Table 3. Top 10 institutions conducting research in the field of new energy vehicle fires, 2019–2024.

Rank	Institution	Country	Total Connections	Number of Publications	Total Citations	Average Citations
1	University of Science and Technology of China	China	127	167	5962	35.7
2	Tsinghua University	China	75	90	4727	52.5
3	Nanjing Tech University	China	33	49	898	18.3
4	Beijing Institute of Technology	China	33	45	1261	28.0
5	Chinese Academy of Sciences	China	27	33	1405	42.6
6	Jiangsu University	China	38	27	657	24.3
7	China People’s Police University	China	41	22	679	30.9
8	City University of Hong Kong	China	27	22	682	31.0
9	China University of Mining and Technology	China	13	20	158	7.9
10	The Hong Kong Polytechnic University	China	15	20	797	39.9

provides information on the top ten institutions in terms of the number of published articles, and Figure 5 shows the inter-institutional collaboration. The country of the institution, the total number of connections, the total number of citations, and the average number of citations are also presented in Table 3. China was the dominant force in the area of NEV fires associated with research on LIB systems, as evidenced by the fact that all ten of the top institutes in terms of publications were from China. The bibliometric analysis revealed that three institutions dominated scholarly output within the selected group, with the University of Science and Technology of China (167), followed by Tsinghua University (90) and Nanjing Tech University (49). The parameter reflecting the cooperation relationship of an institution was the total number of links; the more the number, the stronger the strength of the connection of the institution. The institutions with the highest number of connections were the University of Science and Technology of China (127), Tsinghua University (75), and China People’s Police University (41). Although Chinese institutions entered this research field later than their counterparts in Europe and the United States, the publication volume has grown rapidly in recent years. This growth is driven by the country’s large population, the surge in NEV sales, and increasing concerns over safety issues such as battery fires. However, China is still in a learning phase, drawing on the experiences of early-developing nations, and a gap remains compared to the world’s leading research institutions. With strong market potential and continued economic growth, China is expected to play a central role in future research on NEV fire safety.

With the use of VOSviewer software, a collaborative network graph of 52 major research institutions was generated (Figure 5). The graph’s node sizes represent the volume of published research, the connecting lines show collaboration between two institutions, line widths show the strength of the cooperation, and the color shows the cluster created by the institutions that teamed up more closely. The network mapping revealed five institutional clusters differentiated by collaboration density patterns (Figure 5).

The institutions with the closest cooperation were the University of Science and Technology of China (USTC) and Jiangsu University, with the main research focusing on the TR mechanism and fire prevention and control of batteries [32]. The University of Science and Technology of China and the City University of Hong Kong had the second strongest cooperation, with battery TR characteristics and the development and system design of LIB pack efficient thermal management materials as the main research directions [33,34]. In general, collaboration between research institutions remained relatively fragmented, with most cooperative activities occurring domestically or within nearby regions. This trend hindered the formation of large-scale, interconnected academic networks. The spatial distribution and research priorities appeared to be the dominant factors influencing academic partnerships and communication.

3.2.3. Major Journal Distribution

Journals play a vital role in the dissemination of research findings and the advancement of scientific knowledge [35]. Numerous significant scientific advancements and discoveries are disseminated through journals, and scientific research organizations regard the number of papers published in reputable journals as an important metric for evaluating the outcomes of scientific research. Analyzing journal distribution helps researchers identify the core publication outlets in a given field and access relevant, high-impact references more efficiently. The top ten journals in the field, in terms of the number of publications, along with their detailed data—including the number of papers, the average number of references, the form of journal, and the latest impact factor in 2023—are outlined in

Table 4. Top 10 journals in the field of new energy vehicle fires, 2019–2024.

Rank	Journal Title	Quantity	Average Citations	Citation Indicator	Impact Factor (2023)
1	Journal of Energy Storage	112	14.82	SCIE	8.9
2	Fire Technology	47	18.62	SCIE	2.3
3	Process Safety and Environmental Protection	45	22.51	SCIE	6.9
4	Journal of Power Sources	43	34.88	SCIE	8.1
5	Applied Thermal Engineering	41	26.68	SCIE	6.1
6	Energies	41	10.10	SCIE	3.0
7	Batteries-Basel	39	10.21	SCIE	4.6
8	Chemical Engineering Journal	29	23.90	SCIE	13.3
9	Energy Storage Materials	29	60.14	SCIE	18.9
10	Journal of The Electrochemical Society	24	18.38	SCIE	3.1

. As shown in Table 4, Journal of Energy Storage, Fire Technology, and Process Safety and Environmental Protection were the top three ranked journals in terms of the number of publications in the research field of fires in NEVs, and their corresponding impact factors were 8.9, 2.3, and 6.9. Journal of Energy Storage published 112 papers, accounting for 7.95% of the total number of papers, far more than other journals, and the element of the included papers was heavily related to the safety issues of thermal management of NEV power batteries and their prevention and control. With an average of 60.14 citations, Energy Storage Materials had the most of them, and its impact factor was 18.9, indicating that the journal had a greater influence on the research of thermal stability of NEV power battery materials, electrolyte, and diaphragm material improvement, etc. The relevant articles about NEV power batteries included in the journal were borrowed and researched by many scientists. At the same time, all the related journals were included in the Science Citation Index Expanded (SCIE), indicating that the research in this area was mainly concentrated in the field of natural science.

3.3. Research Knowledge Base

The knowledge foundation of a research field is reflected in journal and literature co-citations. Co-citation refers to the occurrence of two documents being cited together in the reference list of a third document [36]. The core knowledge base of the field comprises these co-cited publications. This knowledge base is further indicated by the journals in which the co-cited literature appears. Frequently co-cited journals typically represent the leading sources at the forefront of research in the field [37]. To identify the knowledge foundation and carriers in the area of NEV fires associated with LIB systems research, the co-cited literature and its sources were analyzed using VOSviewer, Bibliometrix R, and CiteSpace.

3.3.1. Core Author Analysis

Highly productive authors are recognized as leading figures in their fields, whose core research content reflects the fundamental methodologies of the discipline and plays a pivotal role in advancing all aspects of the research process [38]. Author co-occurrence analysis was conducted using CiteSpace based on publications from 2008 to 2024, resulting in a network map comprising 295 nodes and 379 links, with a network density of 0.0087, as illustrated in Figure 6. In the visualization, each node represents an individual author, with the node size corresponding to the number of publications. Lighter node colors indicate more recent co-occurrence, while the width of the purple rings surrounding nodes reflects the level of betweenness centrality. All of the top ten most productive authors in the field of NEV fire research are based in China, indicating the country’s significant academic contribution to this area. This finding aligns with China’s consistently leading position in NEV production and sales, as well as national policies promoting battery safety research. In addition, several authors with high betweenness centrality—such as Feng Xuning, Ouyang Minggao, and Wang Qingsong—demonstrated interdisciplinary collaboration by engaging with multiple research teams, reflecting a robust cooperative network and a strong tendency toward cross-disciplinary integration.

The most prolific author was Wang Qingsong, with a total of 60 published articles. His research primarily focuses on the combustion dynamics of LIB fires and related fire prevention technologies. His collaborative study with Sun systematically revealed the coupling relationship between gas generation and the temperature rise rate during the TR process of LIBs [39]. It also predicted the evolution of internal pressure and fracture behavior under external heating conditions, offering a theoretical basis for fire response modeling and the establishment of safety thresholds. Furthermore, Wang et al. (2023) conducted controlled thermal abuse experiments to compare the TR behavior of large-format LIBs with different chemical compositions [40]. The results demonstrated that lithium iron phosphate batteries exhibited significantly better thermal stability than lithium nickel cobalt manganese oxide batteries, providing valuable insights for battery selection and safety design in the NEV industry. In addition, Chen et al. (2024) carried out innovative research on quasi-solid-state electrolytes, emphasizing that improvements in thermal tolerance of electrolytes could enhance the safety of lithium metal batteries [41]. These findings offer meaningful guidance for the design of safety in current battery systems.

Wang Jian’s research has also been primarily devoted to the prevention and mitigation of LIB fires. In collaboration with Ouyang and Chen, Wang developed a tubular composite phase change material (CPCM) battery module with enhanced flame-retardant properties [42]. The incorporation of aluminum hydroxide as a flame retardant significantly reduced the heat release rate, while the optimized tubular structure improved the TR factor of safety (FOS) compared to conventional designs. This study established a novel integration between structural thermal regulation and functional materials, highlighting the growing trend toward system-level thermal protection strategies. In another study, Wang et al. (2024) conducted a comprehensive investigation into the combined influence of charging rate and external heat flux on TR behavior [43]. A semi-empirical ignition model was proposed, and the findings demonstrated that higher charging states could accelerate the onset of TR, offering valuable insights for optimizing battery management systems. Furthermore, Wang Jian has actively engaged in interdisciplinary collaboration with research teams in materials and safety engineering, promoting the practical application of advanced thermal management materials in EV battery systems.

Feng Xuning has primarily concentrated on the integration, optimization, and safety management of power battery systems in NEVs. Ouyang et al. (2019) conducted a systematic comparison of the effects of battery aging on TR behavior [44]. Their findings indicated that the degradation of cathode thermal stability is one of the key drivers of TR, while the deposition of metallic lithium during aging was identified as a major contributor to the intensification of this phenomenon. The study also proposed targeted safety strategies addressing aging-related risks. Building on this, Feng et al. (2024) developed an automated framework capable of identifying the thermal conductivity of battery materials before and after TR [45]. Validated through experimental results, this method effectively quantified changes in thermal diffusivity, offering a novel approach for real-time fire risk monitoring and parameter acquisition for thermal modeling. Feng Xuning and his research team exemplify interdisciplinary collaboration across energy engineering, electrochemistry, and cyber–physical systems, making them a leading force in the field of battery system safety management for NEVs.

In summary, the most prolific authors in the field of NEV fire research have not only achieved significant breakthroughs within their respective areas of expertise but have also advanced collaborative efforts across disciplines. Their cross-domain contributions have collectively driven progress in multiple subfields, including TR mechanisms, protective material design, and system-level safety strategies. Despite these achievements, several critical gaps remain. For instance, greater emphasis should be placed on CFD-based modeling of thermal runaway, which can more accurately capture coupled phenomena such as heat transfer, gas release, and flame spread in confined environments. In addition, system-level safety integration remains limited. Studies should prioritize the development of holistic safety frameworks that integrate battery design, vehicle architecture, thermal management, and emergency response. Early detection technologies also require improvement. Future research should therefore explore multimodal sensing strategies that integrate temperature, gas composition, acoustic, and electrochemical signals with AI-driven data fusion to enable real-time and reliable detection of abnormal conditions. Similarly, while novel suppression methods have been proposed, many remain impractical for large-scale applications due to limitations in scalability, cost, or adaptability. Intelligent suppression agents capable of rapid cooling, toxicity mitigation, and re-ignition prevention, together with hybrid solutions tailored to diverse battery chemistries and vehicle architectures, warrant further exploration. Battery technology innovation and vehicle safety improvements should progress in tandem with fire prevention research. This includes the development of safer electrode materials, flame-retardant electrolytes, and durable pack designs. In parallel, enhancing vehicle protections—such as crashworthiness, structural fire barriers, and automated emergency response systems—is essential for reducing fire risks during accidents or abnormal operating conditions. These efforts lay a solid foundation for the establishment of interdisciplinary research platforms and the development of integrated fire prevention and control systems in the future. By bridging experimental studies, advanced modeling, sensor technologies, suppression strategies, and regulatory frameworks, future work can accelerate the transition from reactive firefighting to proactive fire safety in NEVs.

3.3.2. High-Cited Literature Analysis

Citation analysis is a technique that quantifies the number of citations an article has received to evaluate its academic quality and influence [46]. Highly cited publications often contain representative research findings and valuable content, thereby enhancing their utility as references for other researchers. By analyzing the literature with high citation frequency, it is possible to identify core themes and foundational ideas within a specific research field. In this study, the top ten most cited articles relevant to NEV fires associated with LIB systems were selected for detailed analysis. The total number of citations (STC) was used as the evaluation metric to rank these publications, and the results are presented in Table 5. The table includes the article title, journal name, document type, first author, publication year, STC, institution number, and country number involved. For documents with more than three authors, only the initial author is shown.

As shown by the statistics in Table 5, of the ten most frequently mentioned documents, five were international collaborations and two were collaborations with domestic institutions. This indicates diversity in the study orientations of the organizations within this field and strong partnerships in institutional study. The literature categories of the ten most referenced publications were reviews and articles, with eight being reviews, indicating that reviews were more emblematic as well as had greater reference worth in the study of NEV fires associated with LIB systems. This predominance of review papers indicates that the field of NEV fires associated with LIB systems is still consolidating existing knowledge rather than generating a substantial number of breakthrough experimental results. Therefore, while review studies provide essential synthesis and guidance for the field, there is a clear need for more experimental investigations to advance understanding and drive innovation.

The most cited document was Wang’s [47] ‘A review of LIB failure mechanisms and fire prevention strategies’, published in Progress in Energy and Combustion Science in 2019. With 837 citations, this article not only addressed TR phenomena and related fire behaviors in LIBs but also discussed prevention measures and highlighted future research directions and application prospects. It can be seen that this article was crucial to the discussion of flames in NEVs. The second most frequently cited document was an article about the security of LIBs, published in the Journal of Energy Chemistry by Chen in 2021, with 801 citations. This paper reviewed the operating principle and cell structure of LIBs, described the effects of TR as well as mechanical, electrical, and thermal abuse, analyzed strategies to improve battery safety, and also analyzed safety standards. Additionally, it outlined possible advancements and opportunities for stronger LIBs in the future [48]. ‘Mitigating Thermal Runaway of Lithium-Ion Batteries’ published by Feng in Joule, in 2020 (cited 712), which was in the third place of citations, summarized mitigation methods for TR in LIBs that operated at the component, cell, and system levels, explained the connection among anticipated TR transmission and accidental burning paths, helped to elucidate the connection among TR as well as fires, and also summarized conditions of abuse that could trigger TR [49]. In other articles, Zhao et al. (2021) reviewed the importance of battery thermal management systems (BTMSs) in electric and hybrid vehicles [50]. They discussed the battery heating mechanisms and their impacts on performance. The study also summarized basic air-cooled BTMS designs. Their findings indicated that air-cooling efficiency could be significantly enhanced through innovative battery pack structures, improved cooling channel configurations, and advanced heat-conducting materials. Finally, the authors suggested future research directions for optimizing air-cooled BTMSs in the EV sector. Wang et al. (2020) provided a comprehensive overview of solid polymer electrolytes for LIBs [51]. The study examined ion transport mechanisms and key performance requirements. It also introduced the classification and fabrication methods of polymer electrolytes. Furthermore, the paper described the formation principles of electrode–electrolyte interfaces and discussed the role of composite electrolytes. Finally, it assessed the current development status and future prospects of polymer electrolyte technologies.

Table 5. Top 10 most cited papers in the field of new energy vehicle fires, 2019–2024.

Rank	Reference Number	STC	Title	Authors	Journal	Type	Year	IN	CN
1	[47]	837	A Review of Lithium-Ion Battery Failure Mechanisms and Fire Prevention Strategies	Wang, Q.S.; Mao, B.B.; Stoliarov, S.I.; Sun, J.H.	Progress in Energy and Combustion Science	Review	2019	2	2
2	[48]	801	A Review of Lithium-Ion Battery Safety Concerns: The Issues, Strategies, and Testing Standards	Chen, Y.Q.; Kang, Y.Q.; Zhao, Y.; et al.	Journal of Energy Chemistry	Review	2021	5	3
3	[49]	712	Mitigating Thermal Runaway of Lithium-Ion Batteries	Feng, X.N.; Ren, D.S.; He, X.M.; Ouyang, M.G.	Joule	Review	2020	1	1
4	[52]	426	Safety Issues and Mechanisms of Lithium-Ion Battery Cell Upon Mechanical Abusive Loading: A Review	Liu, B.H.; Jia, Y.K.; Yuan, C.H.; et al.	Energy Storage Materials	Review	2020	1	1
5	[53]	360	A Review of Battery Fires in Electric Vehicles	Sun, P.Y.; Bisschop, R.; Niu, H.H.; Huang, X.Y.	Fire Technology	Review	2020	3	2
6	[54]	288	A Review of Safety Strategies of a Li-Ion Battery	Chombo, P.V.; Laoonual, Y.	Journal of Power Sources	Review	2020	1	1
7	[55]	275	Conductivity and Lithiophilicity Gradients Guide Lithium Deposition to Mitigate Short Circuits	Pu, J.; Li, J.C.; Zhang, K.; et al.	Nature Communications	Article	2019	5	2
8	[56]	265	Environmental Impacts, Pollution Sources, and Pathways of Spent Lithium-Ion Batteries	Mrozik, W.; Rajaeifar, M.A.; Heidrich, O.; Christensen, P.	Energy & Environmental Science	Review	2021	2	1
9	[50]	221	A Review of Air-Cooling Battery Thermal Management Systems for Electric and Hybrid Electric Vehicles	Zhao, G.; Wang, X.L.; Negnevitsky, M.; Zhang, H.Y.	Journal of Power Sources	Review	2021	2	2
10	[51]	206	Reviewing the Current Status and Development of Polymer Electrolytes for Solid-State Lithium Batteries	Wang, H.C.; Sheng, L.; Yasin, G.; et al.	Energy Storage Materials	Article	2020	2	1

STC, sum of the times cited; IN, institution number; CN, country number.

Table 5. Top 10 most cited papers in the field of new energy vehicle fires, 2019–2024.

Rank	Reference Number	STC	Title	Authors	Journal	Type	Year	IN	CN
1	[47]	837	A Review of Lithium-Ion Battery Failure Mechanisms and Fire Prevention Strategies	Wang, Q.S.; Mao, B.B.; Stoliarov, S.I.; Sun, J.H.	Progress in Energy and Combustion Science	Review	2019	2	2
2	[48]	801	A Review of Lithium-Ion Battery Safety Concerns: The Issues, Strategies, and Testing Standards	Chen, Y.Q.; Kang, Y.Q.; Zhao, Y.; et al.	Journal of Energy Chemistry	Review	2021	5	3
3	[49]	712	Mitigating Thermal Runaway of Lithium-Ion Batteries	Feng, X.N.; Ren, D.S.; He, X.M.; Ouyang, M.G.	Joule	Review	2020	1	1
4	[52]	426	Safety Issues and Mechanisms of Lithium-Ion Battery Cell Upon Mechanical Abusive Loading: A Review	Liu, B.H.; Jia, Y.K.; Yuan, C.H.; et al.	Energy Storage Materials	Review	2020	1	1
5	[53]	360	A Review of Battery Fires in Electric Vehicles	Sun, P.Y.; Bisschop, R.; Niu, H.H.; Huang, X.Y.	Fire Technology	Review	2020	3	2
6	[54]	288	A Review of Safety Strategies of a Li-Ion Battery	Chombo, P.V.; Laoonual, Y.	Journal of Power Sources	Review	2020	1	1
7	[55]	275	Conductivity and Lithiophilicity Gradients Guide Lithium Deposition to Mitigate Short Circuits	Pu, J.; Li, J.C.; Zhang, K.; et al.	Nature Communications	Article	2019	5	2
8	[56]	265	Environmental Impacts, Pollution Sources, and Pathways of Spent Lithium-Ion Batteries	Mrozik, W.; Rajaeifar, M.A.; Heidrich, O.; Christensen, P.	Energy & Environmental Science	Review	2021	2	1
9	[50]	221	A Review of Air-Cooling Battery Thermal Management Systems for Electric and Hybrid Electric Vehicles	Zhao, G.; Wang, X.L.; Negnevitsky, M.; Zhang, H.Y.	Journal of Power Sources	Review	2021	2	2
10	[51]	206	Reviewing the Current Status and Development of Polymer Electrolytes for Solid-State Lithium Batteries	Wang, H.C.; Sheng, L.; Yasin, G.; et al.	Energy Storage Materials	Article	2020	2	1

STC, sum of the times cited; IN, institution number; CN, country number.

Based on co-citation analysis using VOSviewer, studies with fewer than 95 co-citations were excluded, resulting in a co-citation network diagram comprising 52 nodes (Figure 7). The dimension of the nodes indicates the number of citations of the co-cited research, and the separation among the nodes represents the intensity of the association among studies; the shorter the gap is, the greater the connection between the co-cited research. Citation frequency is widely accepted as an indicator of a paper’s impact, and the co-citation of papers containing different keywords reflects the influence of specific research themes, thereby helping to determine the distribution of influence across NEV fires associated with LIB systems research topics. It can be observed from Figure 7 that the knowledge base of literature citations in the area of NEV fires during the period of 2019 to 2024 mainly consisted of six categories: among them, the red cluster contained the largest number of nodes (8), followed by the green cluster (7), the blue cluster (6), and the yellow cluster (5). The purple and light blue clusters each contained 4 nodes. The yellow, purple, and light blue clusters were categorized as smaller clusters.

The main clusters were analyzed: The red cluster included Wang’s review on the failure mechanism and fire prevention strategy of LIBs published in 2019 as the main article, which was related to all other clusters [46]. In the same year, Ouyang et al. (2019) also provided a thorough analysis of the thermal risks associated with LIBs and the appropriate safeguards [57]. On this basis, Wang et al. (2021) compared the large-size LIBs with different cathode materials, reviewed the overcharging behavior and TR features of these batteries, and proposed a safety evaluation scoring system to gauge the batteries’ security, which guided engineers to rationally select battery materials in automotive applications [58]. A review of mitigation techniques for LIB TR, published by Feng in 2020, was the core article in the green cluster [49]. Huang et al. (2021) answered some frequently asked questions about battery safety issues that the general public finds confusing, such as EV collisions, LIB security, current security technologies, and solid-state batteries, elucidated the failure mechanisms of LIBs, discussed the reasons behind battery security incidents, and provided recommendations on safeguards to keep battery systems safer [59]. In the blue cluster, Sun et al. (2020) presented the newest fire security problems of EVs pertaining to TR and LIB fire, reviewed examples of battery fires, analyzed the main features of battery fires, and found that the heat release and danger of EV fires were comparable to those of fossil-fueled vehicle fires, and that fire extinguishing faced many challenges [53]. Diaz et al. (2020) conducted the first comprehensive analysis of safety concerns in the LIB industry across multiple sectors [60]. They compared these concerns with the focus of academic research in related publications. The study revealed that most academic efforts emphasized TR and the development of preventive technologies. In contrast, issues such as module and pack-level safety and other fire protection layers have received comparatively little attention. To address this gap, the authors recommend promoting cross-sectoral improvements, integration, and standardization in LIB safety strategies.

3.3.3. Highly Co-Cited Journals Analysis

Co-cited journal analysis is a quantitative method widely employed in bibliometrics and scientometrics and has been applied by researchers across various disciplines both domestically and internationally. This technique enables the identification and classification of journals, helps determine whether a journal occupies a central or peripheral position within a field, and supports the evaluation of academic journal impact. In this study, VOSviewer was used to perform co-cited journal network analysis by selecting the “Co-citation” type and the “Cited Sources” node. After filtering more than 160 initial nodes, a co-citation network consisting of 65 nodes was obtained (Figure 8). The extent of the nodes shows the number of citations of the co-cited journals, and the gap among the nodes of the nodes denotes the intensity of the association among the journals; the shorter the gap, the greater the co-citation connection among the journals. Based on clustering analysis, the 65 journals were categorized into three major clusters: red, green, and blue.

The red clusters were energy and materials science journals dominated by Applied Energy (1726 citations), Energy Storage Materials (1485 citations), and Electrochimica Acta (1131 citations), while the green clusters were Journals of Power Sources (8478 citations), Journal of The Electrochemical Society (2933 citations), and Journal of Energy Storage (1694 citations) as the core of energy and electrochemical journals; the red cluster exhibited the most citations, suggesting that study on NEV fires was primarily focused on energy and materials. This dominance highlights that NEV fire research has been strongly driven by materials innovation and energy storage performance, which translates directly into engineering needs such as the development of thermally stable electrodes, safer electrolytes, and functional separators to mitigate TR risks.

As shown in Figure 8, most journals in the red cluster concentrated on energy storage and conversion, as well as the development and application of novel materials. In contrast, the green cluster was predominantly associated with electrochemical research, batteries, and other energy storage systems, as well as thermal engineering and thermal management technologies. The prominence of these journals reflects a strong engineering orientation toward battery design optimization, pack-level thermal management, and system integration, indicating that advances at the electrochemical level are rapidly translated into practical NEV safety strategies. These journals in the green cluster were also the most closely linked to the other clusters. With the growing emphasis on environmental safety, the number of journals classified under environmental protection has gradually increased; however, a strong interconnection among them has yet to be established. This emerging trend suggests that future NEV fire safety strategies will increasingly integrate environmental considerations, encouraging engineers to adopt greener flame-retardant additives and recyclable safety materials. Given the complexity of fire-related research in NEVs and the ongoing shift toward multidisciplinary integration, it was anticipated that future core journals in this field would gradually evolve from narrowly focused technical publications to more comprehensive and interdisciplinary outlets. NEV fire safety research is moving toward cross-disciplinary solutions—combining material science, electrochemistry, thermal management, and environmental engineering—to address real-world challenges of vehicle safety and sustainability.

3.4. Research Evolution and Hotspots Analysis

3.4.1. Keyword Co-Occurrence and Safety Themes

One way to interpret keywords is as an overview of an article’s primary points. Research hotspots in the field are often reflected in the timing, frequency, and fluctuation patterns of keyword occurrences [61]. A VOSviewer analysis of NEV fire research associated with LIB systems from 2019 to 2024 identified a total of 4262 keywords. Keywords that appeared at least 20 times were selected and analyzed individually, and after appropriate refinement, 20 representative keywords were extracted to generate

Table 6. Top 20 keywords for research in the field of new energy vehicle fires, 2019–2024.

Rank	Keywords	Occurrences	Total Link Strength	Rank	Keywords	Occurrences	Total Link Strength
1	lithium-ion battery	832	3294	11	thermal management system	129	599
2	thermal runaway	433	2101	12	hazards	126	754
3	safety	310	1490	13	electric vehicle	110	532
4	fire	259	1365	14	failure	94	558
5	electrochemical performance	210	818	15	model	91	487
6	behavior	200	1107	16	overcharge	73	432
7	mechanism	181	953	17	anode	66	201
8	thermal stability	171	873	18	abuse	64	401
9	electrolyte	168	849	19	temperature	63	342
10	propagation	161	914	20	internal short-circuit	61	277

. Among the analyzed keywords, ‘lithium-ion battery’ showed the highest frequency, appearing 832 times. The keywords ‘lithium-ion battery’, ‘thermal runaway’, and ‘safety’ were the three most frequently co-occurring terms, with 3514, 2228, and 1596 co-occurrences, respectively. These results indicated that LIBs were the primary focus of research on fire safety in NEVs. Particular emphasis was placed on TR in these batteries. Additionally, studies increasingly focused on ‘enhancing thermal stability’, ‘electrolyte composition’, and ‘electrochemical performance’ to mitigate fire risks. Researchers also aimed to improve battery materials and design safer battery systems to enhance the overall performance of power batteries used in NEVs.

Researchers can determine the research frontiers and hotspots in a certain topic by examining the keywords, which reflect the article’s main ideas. In this study, the ‘All Keywords’ option was employed in the co-occurrence analysis module of VOSviewer. After initially identifying 4262 keywords, those with a frequency of at least 20 occurrences were retained. Synonymous or redundant terms were consolidated to produce a refined co-occurrence network comprising 63 distinct keywords, as presented in Figure 9. In this network, the thickness of the lines between nodes reflects the frequency of co-occurrence between keywords, while the node size corresponds to the occurrence frequency of each keyword. As shown in Figure 9, the analysis revealed four major thematic clusters, centered around the terms ‘lithium-ion battery’, ‘thermal runaway’, ‘thermal management system’, and ‘fire suppression’. Among these, the most frequently occurring keywords were ‘lithium-ion battery’, ‘thermal runaway’, ‘safety’, and ‘fire’, highlighting the central focus areas in the field of NEV fire research.

Power battery thermal management and performance optimization (blue cluster): the biggest node in the blue cluster was the node with the keyword ‘thermal management system’, which mainly reflected the research on power battery thermal management for NEVs. Power batteries generated a lot of heat during charging and discharging, and improper thermal management could easily lead to overheating, TR, and even fire, which seriously threatened the safety of vehicles and people. The vehicle’s overall performance could be improved by an efficient thermal management system that would prevent battery overheating and TR, lower the risk of fire, lessen the frequency of fires and other safety incidents, and increase the battery’s energy density, charging and releasing productivity, and cycle duration. Therefore, many researchers have optimized power batteries’ thermal management systems, which has evolved into a popular research direction. For example, Li et al. (2024) proposed an innovative thermal management technique that integrates 3D finned tubes with phase change materials (PCM) [62]. Their study found that heat transfer efficiency increased with greater fin height. However, efficiency first decreased and then increased as fin width and axial spacing changed. They also examined how different PCM liquid-solid ratios influenced the overall heat transfer performance of the system. Yu et al. (2023) proposed a new flame-resistant Composite Phase Change Material (CPCM) comprising paraffin wax, highly dense polyethylene, broadened graphite, ammonia polyphosphate, red phosphorus, as well as zinc oxide, and found that CPCMs with varying amounts of fire retardants had superior fire retardant properties as well as thermal management effectiveness [63]. These findings have direct engineering implications for NEV design: pack-level cooling strategies, coolant-flow, and manufacturability considerations should be aligned with BMS thermal-control logic, and proposed thermal-management concepts should undergo combined CFD and module/pack-scale validation to confirm performance and safety under abuse scenarios.

Power battery materials and safety performance (red cluster): with a LIB at its core, this cluster spread to other keywords, with “safety”, “electrochemical performance”, “thermal stability”, “electrolyte”, and other terms having higher heat. The primary research item in the study of NEV fires was the LIB. The overall electrochemical performance of power batteries was improved by exploring and researching new electrode materials, optimizing electrolytes, and improving battery structure design. Attention was also given to the balance between the battery’s electrochemical efficiency and thermal stability to ensure that battery safety was not sacrificed while improving the performance. For example, Li, Wang, and Li (2022) proposed the use of temperature-sensitive conductive polymer-based materials as cathode materials [64]. These materials demonstrated excellent electrochemical performance under normal operating conditions. At elevated temperatures, they provided effective self-protection by triggering early warnings without causing a voltage rise during charging. During discharging, the system could be rapidly shut down to prevent further heat accumulation. This approach effectively helped to prevent TR of the battery. Liu et al. (2023) measured the mass shift, heat production, as well as gas release features of three LIBs under various conditions utilizing a conical calorimeter, and further analyzed the differences in the TR actions of the three batteries by means of principal component analysis (PCA) [65]. From an engineering perspective, advances in separators, electrolytes, and flame-retardant additives should be translated into standardized material-screening workflows and subsequently validated at module and pack levels to balance electrochemical performance with thermal and fire safety. Such stepwise validation will facilitate material selection for production.

Battery thermal runaway and fire risk assessment (green cluster): The terms ‘thermal runaway’, ‘propagation’, ‘behavior’, ‘fire’, and ‘mechanism’ were more common in the green cluster. This cluster reflected the study on the mechanism and characteristics of battery TR. Based on experimental data and theoretical analysis, it established the battery TR and fire risk assessment model, offered recommendations for secure building and use of batteries, and improved the security of NEV power batteries. For example, Cho, Park, and Kim (2022) combined electrochemical theory with real-time electrical measurement factor data to propose a novel approach for real-time fire risk assessment during battery pack performance, and experimentally verified the feasibility of the method [15]. Tong et al. (2024) utilized computational fluid dynamics (CFD) to recover gas diffusion prior to a fire or explosion, established exhaust parameters versus time, and provided fire dynamics boundary circumstances for TR in LIBs [66]. Mechanistic and propagation models of TR should be incorporated into battery-pack layout and containment design; engineering practice would benefit from using CFD scenarios to define representative full-scale test conditions and to support pack-level mitigation strategies.

Fire prevention and control technology (yellow cluster): The most frequent keyword in the yellow clusters was ‘water mist’, and the keywords that co-occurred with this keyword included ‘fire suppression’ and ‘efficiency’. This cluster reflects research on efficient fire prevention as well as control technologies. In response to actual fire accidents, the efficiency and effectiveness of fire suppression in NEVs have been continuously improved through the study of advanced fire extinguishing technologies such as fine water mist. Moreover, safety solutions for EVs are shifting from theoretical research to engineering-oriented practical implementations. Guo et al. (2024) discovered that a microencapsulated extinguishing agent enabled in-situ fire suppression for LIBs [67]. This innovation effectively suppressed flames from ternary LIBs and maintained temperatures below 130 °C. The agent could autonomously detect and respond to the early stages of TR. It acted directly on the faulty cell within the battery pack, successfully preventing the escalation and spread of TR. Hu et al. (2024) comprehensively analyzed the evolution of TR and elucidated the ability of fine water mist the suppress TR in LIBs [11]. Comparative assessments of suppression agents should be extended to full-pack and vehicle-scale experiments to evaluate cooling, re-ignition risk, and environmental impact, and to inform practical onboard suppression deployment and automatic-trigger thresholds.

3.4.2. Research Frontiers in Battery Fire Safety

After processing the keywords, the Timeline View analysis option in CiteSpace was selected to generate the keyword timeline visualization (Figure 10). This visualization primarily aims to depict the historical progression of research hotspots by presenting the temporal distribution and interrelations of different keyword clusters. In the timeline view, the vertical axis represented various keyword clusters, while the horizontal axis denoted the time span. Each node corresponds to a specific keyword, and co-occurrence relationships between keywords are indicated by connecting lines. The thickness of these lines reflects the strength of the co-occurrence. The size of a node indicates the frequency of keyword occurrence in a given year, while the color depth of the node signifies the intensity of its citation burst—darker colors represent stronger bursts. This structure allowed for the quantitative depiction of thematic evolution and associations within the research field [68]. Nodes with a purple outer ring represented a betweenness centrality of no less than 0.1, indicating that the associated keywords had stronger connectivity with other keywords. The research period covered the years from 2008 to 2024.

As shown in Figure 10, many frequently occurring keywords, such as ‘lithium-ion batteries’, ‘additives’, ‘behavior’, and ‘performance’, were already present as early as 2008. In this timeline visualization, each cluster—or a group of clusters—can be regarded as representing a distinct research frontier in the field of NEV fire studies. Clusters #1 (Thermal Management), #2 (Failure Mode Mechanisms), #3 (Propagation Prevention), #4 (Flame Retardant), #5 (Risk Assessment), and #9 (Thermal Runaway) primarily focus on the causes of thermal incidents in power batteries, strategies for fire prevention and control, and system-level safety design. Collectively, these clusters constitute the central focus of research on NEV fire safety. The remaining seven clusters, although not directly targeting fire control in NEVs, contribute valuable support in areas such as understanding material safety characteristics, developing multiphysics models, and enabling early fault detection, making them indispensable components of the broader NEV fire research landscape.

To further uncover emerging trends in the field of NEV fire research, this study conducted a burst keyword analysis using CiteSpace, identifying the top 32 keywords with the strongest citation bursts between 2008 and 2024 (as shown in

Table 7. Top 32 keywords with the strongest citation bursts in new energy vehicle fires research.

Keywords	Year	Strength	Begin	End	2008–2024
additives	2008	6.12	2008	2018	▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂
electrochemical property	2009	6.63	2009	2016	▂▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂
electrochemical performance	2009	5.03	2009	2016	▂▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂
cathode material	2009	4.35	2009	2014	▂▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂
trimethyl phosphate	2009	3.77	2009	2020	▂▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂
electrodes	2009	3.41	2009	2019	▂▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂
phosphate	2010	3.84	2010	2015	▂▂▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂
high power	2011	7.36	2011	2020	▂▂▂▃▃▃▃▃▃▃▃▃▃▂▂▂▂
performance	2008	6.24	2012	2015	▂▂▂▂▃▃▃▃▂▂▂▂▂▂▂▂▂
li-ion batteries	2012	4.13	2012	2019	▂▂▂▂▃▃▃▃▃▃▃▃▂▂▂▂▂
thermal stability	2013	6.91	2013	2017	▂▂▂▂▂▃▃▃▃▃▂▂▂▂▂▂▂
electrolytes	2014	10.12	2014	2020	▂▂▂▂▂▂▃▃▃▃▃▃▃▂▂▂▂
safety	2008	5.55	2014	2018	▂▂▂▂▂▂▃▃▃▃▃▂▂▂▂▂▂
explosion	2014	3.84	2014	2015	▂▂▂▂▂▂▃▃▂▂▂▂▂▂▂▂▂
heat release rate	2015	4.6	2015	2017	▂▂▂▂▂▂▂▃▃▃▂▂▂▂▂▂▂
lithium ion battery	2009	4.12	2015	2016	▂▂▂▂▂▂▂▃▃▂▂▂▂▂▂▂▂
calorimetry	2008	3.34	2015	2018	▂▂▂▂▂▂▂▃▃▃▃▂▂▂▂▂▂
graphite	2015	2.91	2015	2016	▂▂▂▂▂▂▂▃▃▂▂▂▂▂▂▂▂
cell	2016	5.24	2016	2019	▂▂▂▂▂▂▂▂▃▃▃▃▂▂▂▂▂
mechanical property	2016	3.77	2016	2021	▂▂▂▂▂▂▂▂▃▃▃▃▃▃▂▂▂
lipf6	2016	3.45	2016	2020	▂▂▂▂▂▂▂▂▃▃▃▃▃▂▂▂▂
fire hazard	2016	3.27	2016	2017	▂▂▂▂▂▂▂▂▃▃▂▂▂▂▂▂▂
challenges	2017	4.24	2017	2019	▂▂▂▂▂▂▂▂▂▃▃▃▂▂▂▂▂
li ion	2016	4.3	2018	2020	▂▂▂▂▂▂▂▂▂▂▃▃▃▂▂▂▂
membrane	2019	4.68	2019	2020	▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂▂
discharge	2019	3.91	2019	2021	▂▂▂▂▂▂▂▂▂▂▂▃▃▃▂▂▂
heat release	2019	3.84	2019	2020	▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂▂
lithium ion	2019	3.05	2019	2020	▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂▂
short circuit	2020	4.09	2020	2021	▂▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂
lithium ion battery safety	2018	3.76	2020	2021	▂▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂
management system	2017	3.26	2020	2021	▂▂▂▂▂▂▂▂▂▂▂▂▃▃▂▂▂
impact	2020	2.99	2020	2022	▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▂▂

‘Year’ indicates the first occurrence of the keyword; ‘Strength’ reflects the intensity of the burst; ‘Begin’ and ‘End’ denote the time span of the burst period. The red segments in the timeline highlight the active burst duration, while the green segments indicate periods without a burst.

). The color histogram in the last column of Table 7 shows the time distribution of keyword citation bursts from 2008 to 2024. The highlighted segment represents the active burst period of each keyword, and the column length reflects the duration of the burst. This technique highlights keywords that have experienced a rapid increase in citation frequency within a short time frame, indicating potential research frontiers. Keywords such as ‘electrolytes’, ‘thermal stability’, and ‘safety’ exhibited prolonged bursts, reflecting the research community’s sustained focus on battery safety issues. In Table 7, terms such as Li-ion batteries, LiPF6, and electrolytes highlight research focusing on cell chemistry and the safety of electrolytes. Keywords including cell, mechanical property, and short circuit point to studies conducted at the cell or module level, closely linked to form factors and potential failure mechanisms. In addition, entries like management system, fire hazard, and heat release rate indicate that the system-level perspective and fire risk evaluations have also been widely addressed. Overall, the burst keyword analysis captures not only the temporal evolution but also specific dimensions of battery fire safety.

Based on Figure 10 and Table 7, the average occurrence time of keywords in the field of NEV fire research was divided into three distinct stages. The first stage, from 2008 to 2012, was characterized by keywords such as ‘lithium-ion battery’, ‘electrochemical property’, ‘high power’, and ‘additives’ reflecting a primary focus on technical performance and fundamental materials. This phase marked the beginning of research in this domain and laid a solid foundation for subsequent studies. The second stage, from 2013 to 2018, saw the emergence of the field, with growing attention toward keywords associated with the causes of NEV fires, such as ‘thermal runaway’, ‘internal short circuit’, ‘abuse’, ‘explosion’, and ‘calorimetry’. Research during this period increasingly investigated TR and accident mechanisms. In response to a rising number of fire incidents involving NEVs, scholars began to analyze underlying causes. Due to the relatively late appearance of these keywords, their total citation frequency remained relatively low. The third stage, from 2019 to 2024, represented a period of rapid development. New keywords such as ‘water mist’, ‘phase change materials’, ‘thermal management’, ‘short circuit’, and ‘membrane’ emerged, indicating a shift in research focus toward efficient battery thermal management, real-time monitoring, and systemic fire prevention strategies. These developments closely align with real-world fire safety challenges in NEVs. For example, frequent fire accidents caused by internal short circuits and unstable electrolytes accelerated interest in thermal stability, flame-retardant materials, and advanced separator technologies. In recent years, the rise of terms such as ‘battery management system’ and ‘impact’ has also reflected a growing need for safety assessment and control throughout the entire battery life cycle, highlighting the importance of real-time detection and fire risk mitigation. Researchers gradually shifted from passive cause identification and mechanism analysis to proactive strategies for thermal control, early warning, and integrated fire safety systems.

Drawing on the data presented in Figure 10 and Table 7, the key research frontiers in NEV fire studies were analyzed, as outlined below.

Clusters #2 (Failure Mode Mechanisms), #3 (Propagation Prevention), and #9 (Thermal Runaway) collectively represented the research frontier concerning the mechanisms and propagation of TR in the context of NEV fires. TR in batteries has been widely recognized as one of the primary causes of fire incidents in EVs due to its sudden onset characteristics. The causes of TR are multifaceted, and once initiated, it can result in combustion and explosion hazards, posing significant safety risks. As such, the mechanisms and propagation paths of TR have long drawn sustained attention from researchers worldwide. Wang et al. (2012) reviewed the mechanisms, fundamental reactions, and experimental studies related to LIB TR, emphasizing that the uncontrolled exothermic reactions among internal materials were the core drivers of fire and explosion events [69]. Feng et al. (2018), through an analysis of representative EV accidents, systematically summarized the triggering conditions and chain reaction processes of TR [70]. A quantitative model for the chain reactions was proposed, along with a three-level protection strategy. In experimental studies, Li et al. (2019) employed accelerated calorimetry to investigate the TR evolution of high-capacity batteries under different states of charge (SOC), offering critical data to support battery module protection design [71]. Furthermore, Feng et al. (2024) uncovered the pivotal roles of gas diffusion and electrolyte migration in the propagation of TR within blade-type batteries [72]. His findings revealed a three-stage propagation mechanism, providing a solid theoretical foundation for the safety design of blade batteries.

Clusters #0 (Thermal Stability), #4 (Flame Retardant), #6 (Advanced Materials), and #7 (Electrolyte Combustion Behavior) reflected the research frontiers of high-safety battery materials and flame-retardant technologies in the context of NEV fire studies. With the continuous growth in the adoption of NEVs and the increasing frequency of TR incidents, there is a growing emphasis on developing safer battery materials and more effective combustion suppression strategies. Researchers have focused on enhancing battery safety through structural innovations, particularly in the design of separators, electrolytes, and flame-retardant materials. Shen et al. (2018) introduced a novel high-safety ceramic separator featuring excellent thermal resistance, flame retardancy, and a wide electrochemical window, along with enhanced ion conductivity and absorption capacity, offering a promising solution for improving overall battery safety [73]. Zou et al. (2021) proposed an additive-free electrolyte with outstanding high-voltage stability, dendrite suppression capability, and long-term cycling performance at low temperatures [74]. By constructing an interface model based on a new solvation structure, the study provided theoretical insights into the relationship between interfacial behavior and electrode performance. From the perspective of material functional integration, Yang et al. (2024) developed a composite phase-change material (PSEMT) that combined phase-change thermal regulation with synergistic flame-retardant functionality [75]. This material significantly improved thermal management and inhibited TR propagation, contributing to a more comprehensive safety framework for battery modules.

Clusters #1 (Thermal Management) and #10 (Thermal Safety) represented the research frontiers of thermal management and thermal safety control in the field of NEV fire studies. With the increasing energy density of NEV batteries, growing attention is being paid to both efficient thermal regulation and safety protection under extreme conditions. Thermal management systems not only serve to dissipate heat during regular operation but also play a vital safety role in preventing and mitigating TR, making them one of the key components in ensuring the stable operation of battery systems. Weng et al. (2022) reviewed the widespread application of phase change material (PCM)-based thermal management systems in energy storage stations and EVs [76]. The study highlighted the dual functionality of these systems in routine heat dissipation and in mitigating thermal hazards under critical failure scenarios. It further compared the thermal stability and flame-retardant performance of modified PCMs, along with structural enhancement strategies, offering important insights into achieving system-level thermal safety and self-protection. Sarvar-Ardeh, Rafee, and Rashidi (2023) investigated, through numerical simulations, the cooling performance of fully porous minichannel cooling plates applied in prismatic battery systems [77]. The results demonstrated that, compared with conventional channels, fully porous microchannels arranged in a parallel flow configuration significantly improved thermal management effectiveness. Li et al. (2024) systematically explored a comprehensive approach to thermal safety control for LIBs [78]. This included TR mechanisms, fire characteristics, conventional and emerging monitoring technologies, thermal management strategies involving various cooling methods, and comparisons of aqueous and non-aqueous extinguishing agents. The study also proposed a conceptual intelligent fire protection system, outlining an integrated solution for LIB thermal safety. Thermal management concepts should be evaluated for manufacturability and lifecycle trade-offs and validated via combined CFD and module/pack-scale testing to ensure acceptable performance, longevity, and safety under both normal and abuse conditions.

Clusters #5 (Risk Assessment), #8 (Electrothermal Modeling), and #11 (Fault Diagnosis) represented the research frontiers of intelligent early warning and fault diagnosis in the study of NEV fires. During real-world operation, battery systems may face various hidden risks such as electrolyte leakage, overheating, and swelling, all of which can trigger TR and even fire incidents. As a result, precise monitoring of battery operating conditions and the development of early fault detection mechanisms have become a major focus in both academic and engineering communities. Ping et al. (2017) developed a three-dimensional electrothermal coupled model based on OpenFOAM, which was capable of simulating current, voltage, and heat evolution throughout the entire process from normal operation to TR in LIBs [79]. The model’s accuracy was validated using experimental data, and a comprehensive analysis was conducted on how factors such as discharge rate, cooling airflow velocity, ambient temperature, and air duct thickness affected battery thermal safety. Zhang et al. (2023) focused on the risk of TR caused by electrolyte leakage in real-world NEVs [16]. By dismantling actual vehicle battery packs and collecting multi-source data—including incremental capacity, voltage variation, and electrochemical impedance spectroscopy (EIS)—a multimodal, multi-classifier fusion fault diagnosis algorithm was constructed based on a cloud computing platform. This approach enabled the identification of failure types and provided early warnings before critical faults occurred. Gan et al. (2024) proposed a flexible pressure sensor that could be integrated between battery modules to monitor cell swelling in real time [80]. This sensor featured high uniformity and sensitivity, making it a practical tool for real-time battery condition monitoring and TR warning. The study offered a new perspective for improving safety management in EVs.

Cluster #12 (Explosion Dynamics and Hazard Control) represented the research frontier of fire suppression and firefighting technologies in the context of NEV fire studies. This area focuses on the fire evolution, explosion mechanisms, and suppression strategies following TR in power batteries, aiming to enhance the emergency response to fire incidents and support the development of safer, more reliable battery systems. Wang et al. (2024) established a TR propagation testing platform based on an actual EV battery pack to systematically evaluate the suppression effectiveness of various firefighting agents [81]. The results showed that perfluorohexanone delayed the propagation of TR by 427 s and significantly reduced cell casing temperatures; however, it was insufficient to completely block propagation between individual cells. In contrast, water mist not only effectively interrupted the spread of TR between cells but also markedly reduced module temperatures, demonstrating excellent cooling and insulation performance. Shan, Zhu, and Wang (2023) conducted overheating-induced TR and gas explosion experiments to examine the composition and explosive limits of gases released during TR in LIBs [82]. The study revealed that under high SOC conditions, battery explosions were more prone to transition into unstable detonation. Based on these findings, a comprehensive evaluation method for assessing TR hazards was proposed. Suppression agents and strategies should be compared in vehicle tests to assess cooling performance, re-ignition potential, thermal impact, and environmental footprint.

3.4.3. Evolution of Research Topics and Safety Strategies

The co-occurrence network-based clustering approach can draw attention to several subjects within the same field. The two dimensions of centrality and density allow the results of this method to be represented graphically in a graph. It can be separated into four quadrants according to centrality and density, each of which denotes a distinct degree of centrality and density. The density metric, which measured the magnitude of the word linkages inside the subject matter, climbed vertically (along the Y-axis), whereas the centrality statistic, which measured the significance of a subject in the evolution of the entire research field examined, grew laterally (along the X-axis). The more study questions corresponded with the cluster, creating a cohesive and comprehensive whole, the stronger these connections were [83]. The number of papers that contained each keyword determined the extent of each area. In addition to being essential to the study network (having strong relationships with the additional themes), the themes in quadrant 1 (upper right) also show an elevated level of growth. The database’s core is represented by these themes. Study themes that had attracted a lot of interest and progressed to a great extent, but were not closely related to other study areas, were referred to as being in quadrant 2 (lower right), which had a lesser degree of thematic centrality and a greater degree of development. Quadrant 3 (top left) demonstrated maturation as an independent research component with low centrality and strong intra-thematic links. It was a weak research region in quadrant 4 (bottom left), where both internal and outside linkages were weak. It is crucial to remember that a map’s nearby themes just show that their centrality and density statistics were related; no links between them were required.

Research on NEV fires associated with LIB systems has exhibited different hotspots over time. According to the three stages of literature dispersion, the subject terms with more than five occurrences were filtered by Bibliometrix R to better show the research hotspots and changes in each phase. From 2008 to 2012, as shown in Figure 11, there were five topics in the first quadrant in this phase. For NEV fires, the main focus was on power battery fires or explosions to carry out research, and there was long-term and stable research on different types of power batteries and electrochemical properties. Research has focused on power battery electrode materials and additives (quadrant 1). Research related to improving the high-performance and safety of power batteries has also continued to develop, with particular focus on batteries designed for high discharge rates and rapid power output, as well as flame-retardant additives, which have attracted significant attention (quadrant 2). The ionization theme was located in quadrant 4, indicating that research on ionization processes in power cells was still marginal (quadrant 4). During the second phase from 2013 to 2018, as illustrated in Figure 12, research on power battery fires concentrated primarily on LIBs. Academic studies have focused on power batteries’ TR and burning characteristics as well as their calorimetric analysis. In recent years, an increasing number of papers have researched the combustion characteristics, electrode materials, electrolyte materials, and properties of power batteries. Carrying on the initial phase’s study theme, the development of high-performance electrode materials for power batteries and the optimization of their electrochemical properties were still hot topics in domestic and international research. Power battery research on fire behavior and smoke properties was developing, and the core theory had not been fully formed (quadrant 2). In quadrant 3, recent research has emerged focusing on two key areas. Firstly, a study was conducted on real-time monitoring and early warning systems for fire risks associated with NEVs. This research employed wireless sensor networks within the fire warning systems of these vehicles to enhance their accuracy and timeliness in detecting fires. Secondly, another study examined the environmental impact and life cycle assessment of NEV fires. This comprehensive analysis evaluated both the fire risks and environmental performance of NEVs throughout their entire life cycle. The findings from this research provided valuable scientific information that could aid in the prevention and control of fires, guide ecological design, and support the sustainable growth of NEVs. Although these subjects had been more extensively studied, the combination of low centrality and high density suggested limited relevance to other research themes. Fire risk and management system of power batteries was a peripheral study theme (quadrant 4), and the related theoretical system was gradually being formed, but it lacked integration with other fields. From 2019 to 2024, as shown in Figure 13, after some time, based on the research on power battery fire characteristics, to optimize power battery performance and improve their safety, the centrality of electrolytes and ionic liquids for power batteries has significantly improved (quadrant 1), which was the main focus of the present investigation in this area. The study of power battery TR and its fire mechanism has comparatively advanced (quadrant 2). In keeping with the initial phase’s study emphasis, the development of flame-retardant materials for power batteries was still one of the focuses of many researchers. In quadrant 3, to solve the safety problems of traditional liquid electrolytes and promote the development of high-energy density and long-life battery technology, the optimization and application research on the efficiency of solid polymer electrolytes for power batteries has received the attention of many researchers. It had low centrality but a high strength of intra-topic connections. The optimization and application for the performance of gel polymer electrolytes for power batteries has received some attention, and the related research is still developing.

4. Conclusions

This study conducted a bibliometric analysis of NEV fire research associated with LIB systems using WoS core data from 2008 to 2024, supported by CiteSpace, VOSviewer, and Bibliometrix R. The study provides an overview of the current status and future directions of NEV fire research by analyzing publication distribution, research activity trends, influential scholars and literature, the underlying knowledge base, and evolving research hotspots across different time periods. The main conclusions are summarized as follows:

• There were three identifiable stages in the development of research on NEV fires: an initial exploratory phase (2008–2012), a period of steady development (2013–2018), and a phase of rapid growth (2019–2024). Foundational studies published during the early years effectively shaped the trajectory for later developments. China emerged as the leading contributor in both output and citations, with China and the US as central hubs in regionally clustered collaborations. Core journals such as the Journal of Energy Storage, Fire Technology, Process Safety, and Environmental Protection indicate that this field possesses an interdisciplinary attribute, encompassing multiple disciplines including thermodynamics, materials science, and electrochemistry. Improving study rigor and citation influence, increasing English and open access dissemination, and fostering multinational institutional partnerships and shared data are all necessary.
• From 2019 to 2024, the knowledge base of research on NEV fires primarily comprised three aspects: core authors, representative literature, and highly cited journals. The top ten most influential authors in the field were identified, all of whom were based in China, highlighting the country’s significant contributions to advancing research in this area. Most of the core literature consisted of review articles, indicating that the field is consolidating existing knowledge rather than producing many breakthrough experiments. Highly co-cited journals could be broadly categorized into two thematic areas: energy and materials science, and energy and electrochemistry. Influential journals, particularly Journal of Power Sources, reflect the focus on battery technology, thermal management, and fire prevention, indicating that the theoretical foundation of this field is now well established.
• The keyword clustering analysis revealed that the primary focus of research on NEV fire was on power battery thermal management and performance optimization, power battery materials and safety performance, TR of batteries, and fire risk assessment, as well as fire prevention and control technology. The research frontiers in the field of NEV fires could be categorized into five major areas: mechanisms and propagation paths of TR, development of high-safety battery materials and flame-retardant technologies, thermal management and thermal safety control, intelligent early warning and fault diagnosis, and fire suppression and firefighting techniques. Research on NEV fires has evolved through three stages: early work (2008–2012) on battery performance and materials, mid-phase studies (2013–2018) on TR mechanisms and accident causes, and recent efforts (2019–2024) emphasizing electrolyte optimization, thermal management, real-time monitoring, and integrated fire prevention. This shift reflects a transition from passive analysis to proactive strategies, supported by extensive experimental and numerical studies, and highlights the growing need for comprehensive safety systems across the battery lifecycle.