Bibliometric analysis method, device and equipment for CCUS field and medium

Abstract

The application discloses a CCUS field bibliometrics analysis method, device, equipment and medium, relates to the technical field of bibliometrics analysis, and comprises the following steps: obtaining CCUS field core literature; processing the CCUS field core literature to obtain to-be-analyzed data; performing visual analysis and processing on the to-be-analyzed data by using VOSviewer software, HistCite software and Biblioshiny software, obtaining a visual atlas corresponding to the to-be-analyzed data, and obtaining a topic clustering result by using a trained LDA topic model; and determining overall situation analysis results and co-occurrence analysis results of CCUS field literature according to the visual atlas and the topic clustering result, so that the research status, research level and research trend of the field can be comprehensively understood.

Description

Technical Field

[0001] This application relates to the field of bibliometric analysis technology, and in particular to a bibliometric analysis method, apparatus, equipment and medium in the field of CCUS. Background Technology

[0002] Over the past few decades, research on carbon capture, utilization, and storage (CCUS) technology has attracted widespread attention from the global academic and industrial communities. Early research primarily focused on exploring the principles, feasibility, and potential role of CCUS technology in addressing climate change. Scientists attempted to determine, through various experiments and theoretical analyses, whether CCUS technology could effectively reduce greenhouse gas emissions and assess its application potential in different industrial sectors. These studies laid the foundation for the development of CCUS technology but also revealed many challenges in its implementation, such as high energy consumption, high cost, and the security of long-term storage.

[0003] As time has progressed, researchers have begun to delve deeper into each aspect of CCUS technology to propose more effective solutions. One key research direction is the optimization and innovation of CCUS technology. Researchers are striving to reduce the energy consumption and cost of CCUS technology and improve its environmental and economic benefits by improving capture techniques, developing new utilization pathways, and enhancing storage efficiency. For example, some research teams are exploring the combination of biomass energy and CCUS technology (BECCS) to achieve negative carbon emissions. Furthermore, the economic assessment and policy support of CCUS technology are also current research hotspots. Researchers are attempting to establish economic models to analyze the feasibility and competitiveness of CCUS technology under different scenarios, providing a scientific basis for policymakers.

[0004] Bibliometric analysis, as an important research method, provides scientific quantitative analysis results based on large amounts of data and statistical models, offering decision-making references. However, current statistical analysis of literature in the CCUS field is insufficient, making it difficult to fully understand the research status and progress, and to obtain the latest research results and perspectives in a timely manner; it is also difficult to assess the quality and impact of academic research, and to identify authoritative and reliable research results; furthermore, the lack of a comprehensive evaluation of the contributions and influence of scholars and research institutions in the field may lead to the overlooking of important research resources and collaboration opportunities. Summary of the Invention

[0005] The purpose of this application is to provide a bibliometric analysis method, device, equipment, and medium in the field of CCUS, which can provide a more comprehensive understanding of the current research status, research level, and research trends in this field, and can more quickly and accurately locate the research results of core research countries and authoritative publishing institutions, effectively solving the current problems of difficulty in grasping the research status, research progress, and accurately determining the research direction in the field of CCUS.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] Firstly, this application provides a bibliometric analysis method in the field of CCUS, including the following steps.

[0008] Obtain core literature in the CCUS field;

[0009] The core literature in the CCUS field is processed to obtain the data to be analyzed;

[0010] The data to be analyzed was visualized using VOSviewer, HistCite, and Biblioshiny software to obtain the corresponding visualization map.

[0011] The trained LDA topic model is used to cluster the data to be analyzed to obtain topic clustering results;

[0012] Based on the visualization map and the topic clustering results, the overall situation analysis results and co-occurrence analysis results of CCUS field literature are determined; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, co-occurring cited literature and co-occurring keywords.

[0013] Secondly, this application provides a bibliometric analysis device in the field of CCUS, including the following modules.

[0014] The CCUS core literature acquisition module is used to acquire core literature in the CCUS field.

[0015] The core literature processing module is used to process the core literature in the CCUS field to obtain the data to be analyzed.

[0016] The visualization analysis module is used to perform visualization analysis on the data to be analyzed using VOSviewer software, HistCite software, and Biblioshiny software to obtain the visualization map corresponding to the data to be analyzed.

[0017] The clustering module is used to cluster the data to be analyzed using a trained LDA topic model to obtain topic clustering results;

[0018] The analysis result determination module is used to determine the overall situation analysis results and co-occurrence analysis results of CCUS field literature based on the visualization map and the topic clustering results; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, the analysis results of co-occurring cited literature and the analysis results of co-occurring keywords.

[0019] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the bibliometric analysis method in the CCUS field described above.

[0020] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the bibliometric analysis method in the CCUS field described above.

[0021] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a bibliometric analysis method, apparatus, equipment, and medium in the CCUS field to obtain core literature in the CCUS field; processes the core literature in the CCUS field to obtain data to be analyzed; uses VOSviewer software, HistCite software, and Biblioshiny software to perform visualization analysis on the data to be analyzed to obtain a visualization map corresponding to the data to be analyzed; uses a trained LDA topic model to cluster the data to be analyzed to obtain topic clustering results; and determines the overall situation analysis results and co-occurrence analysis results of CCUS field literature based on the visualization map and topic clustering results. This application uses VOSviewer software, HistCite software, and Biblioshiny software to perform visualization analysis on the data to be analyzed to obtain a visualization map corresponding to the data to be analyzed, thereby obtaining the overall situation analysis results and co-occurrence analysis results of the literature. This allows for a more comprehensive and systematic understanding of the research status, research level, and research trends in this field, and enables faster and more accurate identification of the research results of core research countries and authoritative publishing institutions. This solves the current problem in CCUS research that it is difficult to grasp the research status, research progress, and accurately determine the research direction. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a diagram illustrating the application environment of a bibliometric analysis method in the field of CCUS, as described in one embodiment of this application.

[0024] Figure 2 This is a flowchart illustrating a bibliometric analysis method in the field of CCUS, provided as an embodiment of this application.

[0025] Figure 3 This is a schematic diagram illustrating the specific process of a bibliometric analysis method in the field of CCUS provided in an embodiment of this application.

[0026] Figure 4 This is a distribution chart of annual publications in the CCUS field provided in an embodiment of this application.

[0027] Figure 5 This is a schematic diagram of the number of documents issued by a major issuing agency according to an embodiment of this application.

[0028] Figure 6 This is a schematic diagram of the number of national documents issued according to an embodiment of this application.

[0029] Figure 7 This is a schematic diagram of the country distribution map of CCUS publications provided in one embodiment of this application.

[0030] Figure 8 This is a schematic diagram of the distribution map of CCUS issuing agencies provided in one embodiment of this application.

[0031] Figure 9 This is a schematic diagram of a network knowledge graph based on co-occurring keywords provided in an embodiment of this application.

[0032] Figure 10 This is a keyword highlighting diagram provided in an embodiment of this application.

[0033] Figure 11 This is a schematic diagram showing the frequency of keyword occurrences provided in an embodiment of this application.

[0034] Figure 12 This is a distribution map of core journals based on Bradford's Law, provided in one embodiment of this application.

[0035] Figure 13This is a keyword development relationship diagram provided in one embodiment of this application, showing the relationship between keywords every five years.

[0036] Figure 14 This is a schematic diagram of the LDA topic model analysis process provided in an embodiment of this application.

[0037] Figure 15 This is a visualization diagram of document topic clustering provided in an embodiment of this application. Detailed Implementation

[0038] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] The bibliometric analysis method in the field of CCUS provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send core CCUS literature to server 104. After receiving the core CCUS literature, server 104 processes it to obtain the data to be analyzed. It then uses VOSviewer, HistCite, and Biblioshiny software to perform visualization analysis on the data, obtaining a visualization map corresponding to the data. Based on the visualization map, it determines the overall situation analysis results and co-occurrence analysis results of CCUS literature. Server 104 can then feed back the obtained overall situation analysis results and co-occurrence analysis results for CCUS literature to terminal 102. In addition, in some embodiments, the bibliometric analysis method in the CCUS field can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly perform bibliometric analysis on the core literature in the CCUS field, or the server 104 can obtain the core literature in the CCUS field from the data storage system and perform bibliometric analysis on the core literature in the CCUS field.

[0041] The terminal 102 can be, but is not limited to, various desktop computers, laptops, and tablets. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0042] In one exemplary embodiment, such as Figure 2 As shown, a bibliometric analysis method in the CCUS field is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 205.

[0043] Step 201: Obtain core literature in the CCUS field.

[0044] Step 202: Process the core literature in the CCUS field to obtain the data to be analyzed.

[0045] Step 203: Use VOSviewer, HistCite, and Biblioshiny software to perform visualization analysis on the data to be analyzed, and obtain the visualization map corresponding to the data to be analyzed.

[0046] Step 204: Using the trained LDA topic model, cluster the data to be analyzed to obtain topic clustering results.

[0047] Step 205: Determine the overall situation analysis results and co-occurrence analysis results of CCUS field literature based on the visualization map and the topic clustering results; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, the analysis results of co-occurring cited literature and the analysis results of co-occurring keywords.

[0048] By implementing steps 201 to 205 above, VOSviewer, HistCite, and Biblioshiny software are used to perform visual analysis of core data from core literature in the CCUS field. Based on the corresponding visualization maps, the overall situation analysis results and co-occurrence analysis results of the literature are obtained. This provides a comprehensive understanding of the research status and level in the CCUS field, enabling faster and more accurate identification of research results from core research countries and authoritative publishing institutions, which is beneficial for practitioners to better establish research directions and priorities. By using HistCite software to draw a network knowledge graph based on co-occurrence of keywords, and using Biblioshiny software to create keyword timeline diagrams and keyword burst diagrams, relevant personnel can better understand the constantly changing research directions and popular research objects in the current CCUS field, providing a reference for establishing research directions and key research areas. This application aims to address the shortcomings in CCUS (Carbon Capture, Utilization, and Storage) research in related technologies by proposing a bibliometric analysis method to comprehensively and systematically analyze the research status in this field, overcoming the current research difficulties in grasping the research status and progress in the CCUS field and in accurately determining research directions.

[0049] like Figure 3 As shown, the bibliometric analysis method in the CCUS field can be divided into four steps: literature extraction, literature screening, bibliometric analysis using multiple bibliometric analysis tools, and network visualization. Literature extraction utilizes the WOS system platform for keyword retrieval to obtain relevant domestic and international CCUS literature, yielding 8650 articles. Then, manual literature screening is performed using a manual selection formula based on title and abstract information to retain high-quality literature, resulting in 6753 core CCUS articles. Multiple bibliometric analysis tools (VOSviewer, HistCite, and Biblioshiny software) are used for bibliometric analysis, and network visualization is performed to obtain a visual map corresponding to the analyzed data, including collaboration status, hot topic trends, statistical subjects, and spatiotemporal distribution information.

[0050] In another exemplary embodiment of this application, step 201 of obtaining core literature in the CCUS field specifically includes: obtaining relevant literature in the CCUS field at home and abroad; and selecting core literature in the CCUS field from relevant literature in the CCUS field at home and abroad.

[0051] The specific process for obtaining relevant domestic and international CCUS literature and selecting core literature in the CCUS field is as follows: On the Web of Science website, select the Web of Science Core Collection database; select the Science Citation Index Expanded—1990—present as the citation index for the Web of Science Core Collection; select the keyword search formula as: TS = (“CCUS” or “carbon capture, utilization and storage” or “carbon capture and storage” or “carbon capture and sequestration” or “CO2 capture, utilization and sequestration” or “CO2 capture, utilization and storage” or “CCS” or “CO2 capture and utilization” or “CO2 capture and storage” or “CO2 capture and storage” Advanced search was performed using "sequestra*(carbon dioxide capture and storage*)" or "carbon capture and utilization(carbon capture and utilization)" as search criteria, with TS representing the search term. The time span was selected as January 1, 2000 to December 10, 2024, the document type was selected as academic papers, and the language was selected as English, which yielded core literature in the CCUS field.

[0052] In step 202 above, the core literature in the CCUS field is processed to obtain the data to be analyzed. Specifically, this includes: creating a new folder CCUS, which includes folders named data, input, output, and project; exporting the core literature in the CCUS field to obtain plain text files of the core literature in the CCUS field; naming the exported plain text files of the core literature in the CCUS field starting with download and saving them to a folder named input to obtain the data to be analyzed.

[0053] Before using VOSviewer, HistCite, and Biblioshiny software to perform visualization analysis on the data to be analyzed and obtain the corresponding visualization map, the bibliometric analysis method in the CCUS field further includes: deduplicating the data to be analyzed to obtain deduplicated data to be analyzed.

[0054] The data to be analyzed is deduplicated to obtain deduplicated data. Specifically, this includes: using deduplication software, selecting folders named input and output in the CCUS folder, and clicking deduplication and sorting; storing the data in the output folder into the data folder to obtain the deduplicated data to be analyzed.

[0055] Specifically, the deduplication process includes: running the relevant software and selecting data input / output; selecting folders named input and output within the CCUS folder; and clicking "Deduplication and Organize"; storing the data from the output folder into a data folder to obtain the deduplicated data to be analyzed. The relevant software refers to VOSviewer, HistCite, and Biblioshiny.

[0056] The data to be analyzed was visualized using VOSviewer, HistCite, and Biblioshiny software to obtain corresponding visualization maps. Based on the visualization maps, the overall situation analysis results and co-occurrence analysis results of the literature were obtained. The overall situation analysis results of the literature included the analysis results of the publication time, publishing institution, and publishing country of the literature. The co-occurrence analysis results of the literature included the analysis results of co-occurring authors, co-occurring cited documents, and co-occurring keywords.

[0057] In step 203 above, the data to be analyzed is visualized using VOSviewer, HistCite, and Biblioshiny software to obtain the visualization map corresponding to the data to be analyzed. Specifically, this includes the following steps 301 to 304.

[0058] Step 301: Draw a line chart based on the year and number of articles published in the data to be analyzed.

[0059] Obtain the data to be analyzed regarding the relevant years and the number of articles published, and draw a line chart, such as... Figure 4As shown in the line graph, the horizontal axis represents the year, and the vertical axis represents the number of publications. The results show that research in the CCUS field started relatively early, and the overall number of publications shows an upward trend. In the first stage (2000-2010), a comparison of the top ten countries in terms of the number of published papers reveals that although research in this field started relatively late in all countries, and the increase in the number of papers was also relatively slow during these ten years, with only 1324 publications, averaging only 14 articles per country per year, indicating that all countries were in the exploratory stage in the CCUS field during this period. The second stage (2010) From 2010 to 2017, the number of publications in CCUS showed an increasing trend worldwide, especially in China, the United States, and the United Kingdom. From 2010 to 2016, the United States consistently published more papers than China. However, with China's increased focus on environmental issues, China surpassed the United States in CCUS publications in 2017. In the third phase (2018-2024), as China placed greater emphasis on engineering research within CCUS, the rate of increase in related papers accelerated significantly, far exceeding that of other countries, indicating that China has achieved significant success in this field.

[0060] Step 302: In VOSviewer software, set the time range to 2000-01-01 to 2024-12-10, select a time span of 1 to 4 years, select the node type as institution, country, co-author, cited documents and keywords in order, select Cosine for link strength, set g-index to 5 in node filtering method, select Pathfinder method, and draw a knowledge graph of literature-related information network.

[0061] Step 303: In HistCite software, perform research trend and frontier evolution analysis on the data to be analyzed, obtain research directions and hot topics in the CCUS field, and draw a network knowledge graph based on co-occurrence of keywords.

[0062] Step 304: Perform keyword co-occurrence analysis and cluster analysis on the data to be analyzed in Biblioshiny software to obtain the frequency of keyword occurrence, betweenness centrality, and related research topics, and draw keyword timeline and keyword burst maps; line graphs, literature-related information network knowledge graphs, keyword co-occurrence-based network knowledge graphs, keyword timeline and keyword burst maps constitute the visualization map corresponding to the data to be analyzed.

[0063] Before determining the overall situation analysis results and co-occurrence analysis results of CCUS literature based on the visualization map and the topic clustering results, the bibliometric analysis method of CCUS also includes: optimizing the visualization map based on the g-index and Pathfinder algorithm to obtain an optimized visualization map.

[0064] In the visualization graph, the research object to be visualized is taken as a node, and the frequency of node occurrence or betweenness centrality is taken as a node attribute; betweenness centrality reflects the extent to which a node can influence other nodes in the network; the strength between nodes is represented by the Cosine algorithm, and the connection relationship between nodes is established.

[0065] This application presents a dynamic optimization method for knowledge graphs based on g-index and the Pathfinder algorithm. It quantifies node importance by constructing a composite g-index model, integrating local influence, global PageRank, and domain prior weights. Combined with a multi-scale Pathfinder algorithm, it dynamically adjusts path thresholds and weight decay rates to achieve network sparsity. A bidirectional verification mechanism is designed to ensure the semantic consistency and topological connectivity of the pruned graph, and an incremental update architecture is established to improve computational efficiency. In CCUS knowledge graph applications, this method significantly improves query efficiency and critical path retention, overcoming the technical shortcomings of traditional methods such as node misjudgment, insufficient handling of redundant edges, and poor dynamic adaptability.

[0066] The composite g-index model in this embodiment compensates for the shortcomings of the traditional g-index by introducing multi-dimensional parameters or mixed indicators. The composite methods include: time-weighted composite, discipline normalization, collaborative contribution weighting, and multi-indicator fusion.

[0067] (1) Time-weighted composite, adjusting the weight of CCUS field literature through a dynamic window: introducing a time decay factor (such as exponential decay), and giving higher weight to recent citations.

[0068]

[0069] Among them, G tw The time-weighted composite weighting of CCUS-related literature; g is the value of the g-index, representing the size threshold of a scholar's high-impact paper collection. The g-index is calculated based on the number of a scholar's publications and citations, and aims to assess a scholar's academic output; C i Let be the number of citations of the i-th paper; λ be the decay rate, and t be the citation count of the i-th paper. i This refers to the number of years in which CCUS-related literature has been published.

[0070] (2) Subject normalization: Adjust the g-index based on the average citation level of the subject. The process is as follows: Calculate the average g-index of a subject and record it as the subject benchmark value. Divide the individual g-index by the subject benchmark value to obtain the normalized composite index. The formula for calculating the normalized composite index is shown in formula (2).

[0071]

[0072] Among them, G norm G is the normalized composite index; G is the individual g-index; G 学科基准 This serves as a benchmark value for the discipline.

[0073] (3) Collaborative contribution weighting.

[0074] Role weighting: Differentiate author contributions (e.g., first author and corresponding author have a weight of 1, other authors have a weight of 0.5).

[0075]

[0076] Among them, G c Weighted contribution to cooperation; w i The author's role weight.

[0077] (4) Integration of multiple indicators.

[0078] Combined with the h-index (an h-index is a hybrid quantitative indicator used to assess the quantity and quality of researchers’ academic output): construct the hg-index to balance quantity and quality.

[0079] hg=α·h+β·g (4).

[0080] Where hg represents hg-index; h represents h-index; g represents g-index; α and β are the weighting coefficients of h-index and g-index, respectively.

[0081] Suppose scholar A is in the field of computer science (G 学科基准 The original individual g-index of the individual (=15) is 20, with 3 out of 5 highly cited papers (highly cited papers are those with a high citation rate) being the first author. If subject normalization + role weighting (i.e., collaborative contribution weighting) is used: the normalized composite index G norm =20 / 15≈1.33. After role weighting, the total citation probability may decrease, leading to G c Slightly below 20, but more in line with actual contributions.

[0082] (1) Definition of Local Influence: The influence of a scholar within a specific academic community, institution, or collaborative network, emphasizing the dissemination and recognition of their work within a local scope. Evaluation dimensions: Intra-institutional citations and local collaborative network strength. Intra-institutional citations: the frequency of citations of a paper within its institution or collaborative network; Local collaborative network strength: the degree of collaboration with scholars from the same institution / region (e.g., number of co-authors, impact of collaborative papers).

[0083] (2) Global PageRank definition: The PageRank algorithm is used to measure the importance of scholars’ achievements in the entire academic network based on the citation network structure of academic papers.

[0084] Core idea: Citations of high-impact papers give cited papers higher weight. The formula for calculating global PageRank is shown in formula (5).

[0085]

[0086] Among them, PR(p i ) represents the global PageRank of the i-th cited paper; d is the damping factor (usually 0.85); p k For p i The cited papers, i.e., paper p. j Cited paper p i ;Pr(p j ) for the paper p j Global PageRank; L(p j ) for the paper p j The number of outgoing chains, i.e., the number of chains in paper p. j The number of other papers cited.

[0087] (3) Definition of prior weights for a particular field: Based on the characteristics of the discipline or the stage of development of the field, a predefined weighted adjustment is made to the benchmark influence of academic achievements. Application scenarios can include compensation for less popular fields and incentives for emerging fields. Compensation for less popular fields: achievements in fields with low citation rates need to have their weight increased (e.g., mathematical theory vs. biomedicine); Incentives for emerging fields: fields that are developing rapidly but have not yet accumulated high citations (high citation rate) are given higher potential scores.

[0088] (4) Combine local influence, global PageRank and domain prior weight into the composite g-index model. The calculation formula is shown in formula (6).

[0089]

[0090] Among them: G 综合 For the final g-index; w local For local weighting coefficients; wPageRank The global weight coefficients (satisfying w) local +w PageRank =1); L i For local impact scoring (such as the percentage of local citations); PR i This represents the global PageRank score. The domain prior weights are (e.g., D = 1.2 for mathematics and D = 0.9 for computer science).

[0091] Assume scholar B has an original individual g-index of 12 in the field of theoretical physics (a niche field, with a field weight D = 1.3); local influence: 50% of his citations come from the same research institute, L i =0.5; Global PageRank: The number of times a major paper is cited by Nobel laureates, PR i The average value is 2.5. Therefore, the final g-index for scholar B, calculated comprehensively, is: After adjustment, the cumulative citations increased significantly, and scholar B's final g-index G... 综合 =15, which more accurately reflects its cross-dimensional contribution.

[0092] Analysis of issuing institutions: In the ranking of Total Connectivity Strength (TLS), the Chinese Academy of Sciences (314) ranks first, followed by Tsinghua University (125), University of Texas at Austin (85), Beijing Institute of Technology (78), and University College London (71). Figure 5 As shown, in terms of publication volume, the Chinese Academy of Sciences (CAS) has the highest number of publications (231), followed by Tsinghua University (125), University of Texas at Austin (85), Beijing Institute of Technology (49), and University College London (46). Specifically, the CAS has close collaborations with institutions such as China University of Petroleum, Southwest Petroleum University, Tsinghua University, University of Edinburgh, and University of Texas at Austin, primarily focusing on domestic collaborations. The University of Texas at Austin, on the other hand, has close collaborations with international academic and research institutions such as the University of Edinburgh, the CAS, China University of Petroleum, and Delft University of Technology, indicating frequent international collaborations in the CCUS field.

[0093] To conduct a more in-depth and scientific review of literature topics and reduce subjective bias in classification, this study combines LDA topic modeling with manual coding. LDA topic modeling, an unsupervised topic modeling method, aims to discover hidden structures and latent semantics in documents. It is a three-layer Bayesian model comprising a document layer, a hidden topic layer, and a word layer. This LDA topic model treats documents as probabilistic mixtures of topics and topics as probabilistic combinations of words. By estimating the probability distributions of document-topic and topic-word pairs, it clusters documents. This probabilistic approach objectively reveals the latent topics of documents, reducing the subjective errors of manual labeling. Subsequently, by selecting the most probable topic words and combining them with manual coding, the keywords extracted by the LDA topic model are summarized into more abstract research topics.

[0094] Using a trained LDA topic model, the data to be analyzed is clustered to obtain topic clustering results, specifically including:

[0095] (1) Perform data preprocessing on the data to be analyzed to obtain a preprocessed data list. Specifically: filter and merge the literature title, keyword and abstract fields; remove punctuation and unnecessary characters; expand the stop word library and remove stop words; use spaCy to load the pre-trained model for word form restoration to obtain a preprocessed data list.

[0096] (2) Topic modeling and clustering, specifically including: establishing an LDA topic model, setting the number of iterations, and training the LDA topic model to obtain a trained LDA topic model; combining the coherence score and perplexity score to determine the optimal number of topics; and determining the topic-word probability distribution and document topic probability distribution based on the probability score function.

[0097] (3) Visualization and manual coding: Specifically, the pyLDAvis library is used for visualization analysis, including drawing distances between topics and interactive visualization analysis; the topic clustering results are manually coded, including research topics, cluster topics and keywords.

[0098] Figure 14This paper demonstrates the clustering process of project knowledge transfer research topics based on the LDA topic model, which mainly includes data collection, data preprocessing, topic modeling and clustering, visualization and manual coding. This application uses the Web of Science core database as its literature source, obtaining relevant literature through retrieval and screening, and exporting TXT text containing basic literature fields. During data preprocessing, a stop word list was added to include literature search terms ("CCUS", "carbon capture, utilization and storage", "carbon capture and storage", "carbon capture and sequestra", etc.) and general descriptive terms (article, results, authors, copyright notices in the abstract, etc.). In topic modeling and clustering, since the number of topics needs to be determined manually, a consensus score and a perplexity score were used to jointly determine the optimal number of topics; a higher consensus score indicates stronger interpretability, and a lower perplexity score indicates a better fit. In topic visualization and manual coding, manual coding was performed manually. For topics with differing results, two teachers were brought in to discuss and analyze until a consensus was reached. Manual coding was also performed based on the distance between the visualized graphs and the semantic connections of keywords between topics to determine the research topics.

[0099] By using LDA topic modeling, four clustering themes for knowledge transfer in projects—chemical engineering, energy and fuels, environmental engineering, and green sustainable development—were identified. Then, using the LDA topic model output as input, Python was used to create a visualization of the literature topic clustering, as shown below. Figure 15 As shown in Table 1, clustering topics are represented by circles. The size of the circle represents the frequency of the topic's occurrence, the shade of gray of the circle represents different research topics, and the physical distance between the circles represents the proximity between the topics. Combining the spectral information and the relationship between topic keywords, the clustering results were manually coded, and four research topics were identified, as shown in Table 1.

[0100] Table 1 shows the four topic clustering results obtained from the LDA topic model.

[0101]

[0102] National document issuance volume, such as Figure 6As shown in the analysis of publishing countries, based on the data on the changing trends of the top ten countries of origin for authors from 2000 to 2024, it can be seen that China, the United States, the United Kingdom, South Korea, Germany, Australia, Japan, Italy, Canada, and India are in a leading position in CCUS research. Chinese authors have the highest number of published papers and the fastest growth rate in the number of published papers in the last ten years, followed by the United States. Furthermore, in 2017, the number of papers published by Chinese authors surpassed that of the United States for the first time, and the growth rate is much higher than that of other countries. In fact, by 2024, the number of papers published by Chinese authors will exceed the combined number of publications by the second-ranked (United States) and third-ranked (United Kingdom).

[0103] The top ten disciplines in terms of publication volume include Engineering Chemistry, Engineering Environment, Environmental Science, Multidisciplinary Chemistry, Engineering Machinery, Energy and Fuel, Green and Sustainable Technology, Engineering Electrical and Electronic, Applied Physics, and Civil Engineering. The percentages of publication volume for the top ten disciplines are as follows: Chemical Engineering 27.86%, Energy and Fuel 19.44%, Environmental Engineering 17.28%, Green and Sustainable Technology 12.1%, Environmental Science 6.98%, Electrical Engineering 6.48%, Multidisciplinary Chemistry 3.89%, Applied Physics 2.09%, Architectural Engineering 2.09%, and Geotechnical Engineering 1.8%.

[0104] CCUS publications include a country distribution map, as shown below. Figure 7 As shown in the figure, the distribution map of CCUS publishing institutions is as follows: Figure 8 As shown, the network knowledge graph based on co-occurrence of keywords is as follows: Figure 9 As shown in the keyword burst graph, Figure 10 As shown, the frequency of keyword occurrence is as follows: Figure 11 As shown.

[0105] Keyword analysis: Keywords with high frequency and middleness centrality, such as carbon capture, utilization, and storage, appear frequently, indicating that these are important terms in CCUS research. Keyword clustering reveals several major research themes, such as carbon capture technology, carbon utilization pathways, and carbon sequestration methods, representing the main research framework in this field.

[0106] This embodiment uses Bradford's Law to define the core area of ​​journals in this field, namely the International Journal of Greenhouse Gas Control, Applied Energy, Journal of Cleaner Production, Fuel, and Chemical Engineering Journal, with a total of 2465 articles. This indicates that these five journals account for 36.5% of the total number of articles published by the 571 journals in the core area. Therefore, in the future, scholars who want to understand the important research directions in CCUS-related fields only need to search the core area journals, avoiding wasting time searching through a large number of documents.

[0107] Bradford's Law's core strength lies in revealing the concentrated patterns in literature distribution through quantitative analysis, enabling the rapid identification of core journal clusters within a specific subject area. Its model for dividing the core, relevant, and irrelevant areas (1:n:n) 2 This principle can guide literature institutions to optimize resource allocation, prioritizing the purchase of high-value core journals with limited funds, and significantly improving the efficiency of information acquisition. It also provides data support for the analysis of disciplinary development trends and the design of literature retrieval systems, combining theoretical rigor with practical guidance.

[0108] The formula for calculating the cumulative number of relevant papers corresponding to the journal ranking number n is shown in formula (7).

[0109]

[0110] Where R(n) is the cumulative number of relevant papers corresponding to n; n is the journal ranking number; α is the number of relevant papers in the first-level journals R(1); c is the total number of journals in the core area; N is the total number of journals in the ranking; β is a parameter related to the number of journals in the core area; K is a coefficient equal to the slope of the straight line part in the dispersion curve; s is a parameter whose value is equal to the value of n at the intersection of the straight line part of the graph extended backward and the horizontal axis. When N is large enough,

[0111] Bradford's dispersion law applies not only to divisions of three zones but also to divisions of more than three zones. That is, regardless of the selected region, the proportional coefficient changes accordingly with different zones. It has also been proven that journals ranked can be divided into any number of zones, yielding an expression similar to Bradford's dispersion region description: n1:n2:n3:…n m =1:a:a 2 :…a m .

[0112] The distribution map of core journals based on Bradford's Law is as follows Figure 12 shown. Bradford's Law effectively identifies the core journal group to focus on high-impact research by revealing the literature distribution law in the bibliometrics of the CCUS field. This law divides journals into a core area, a related area, and a peripheral area (with a ratio of 1:a:a 2 ), which can quickly locate the core journals that centrally publish high-quality literature in the CCUS field (such as energy and environmental journals), reducing the cost of screening a vast amount of literature. At the same time, the analysis of its dispersion law can reflect the characteristics of interdisciplinary research. For example, the literature related to carbon capture technology may be dispersed in multidisciplinary journals such as chemical engineering, geology, and policy research, assisting researchers in systematically tracking cross-field research trends. Combining co-word analysis and citation networks, Bradford's Law can also optimize hotspot identification and trend prediction, providing data support for the CCUS technology route planning.

[0113] Analysis of cited literature: Only 154 papers have been published in the CCUS field, with the local citation times and total citation times being 950 times and 9762 times respectively. However, the average local citation times and total citation times are the highest values, being 6.2 times per paper and 63.4 times per paper respectively, indicating that as an international and interdisciplinary journal, it not only covers fields such as chemical engineering, environment, and sustainable energy development, but also has great influence in the CCUS field. The cited journal frequency data is shown in Table 2.

[0114] Table 2 Cited journal frequency data

[0115]

[0116]

[0117] This application also uses bibliometrics to construct a Sankey diagram for the CCUS field, which can intuitively reveal the discipline dynamics through multi-dimensional data integration. Through the literature-author-keyword Sankey diagram, the core research teams, high-frequency technology topics, and knowledge association paths can be quickly identified to track the evolution of field hotspots and the trend of interdisciplinary integration; the institution-author-citation Sankey diagram can clearly show the institutional contribution, the academic influence of scholars, and the knowledge dissemination network, quantitatively evaluating the scientific research cooperation efficiency and the intensity of achievement transformation. This visualization method effectively reduces the difficulty of analyzing complex relationships, provides data support for scientific research layout, resource optimization, and technology route planning, and has the dual advantages of grasping macro trends and mining micro associations.

[0118] The development relationship of keywords every five years is as follows Figure 13 shown. The Sankey diagram display ability of each piece of information: It can methodically reveal the hierarchical relationship and flow direction between different nodes, helping researchers better understand the complex data flow process.

[0119] Research Trend Analysis: CCUS research can be roughly divided into three stages: ① From 2007 to 2014, the emergence frequency of Integrated Gasification Combined Cycle (IGCC), Carbon Dioxide Capture and Storage (CCS), Coal Gasification, and Oxyfuel was relatively strong, and the emergence time of Carbon Dioxide Capture and Storage and Integrated Gasification Combined Cycle was relatively long. Oxyfuel technology can completely burn fuel into carbon dioxide and water, while simultaneously capturing, concentrating, and storing carbon dioxide to prevent it from being directly released into the atmosphere and causing pollution. Carbon Dioxide Capture and Storage is the process of separating carbon dioxide from atmospheric, industrial, or energy-related emission sources and sealing it in geological reservoirs, thus isolating it from the atmosphere for a long period of time. IGCC technology, through the coal gasification process, can capture and treat carbon in coal before combustion, thereby achieving near-zero carbon dioxide emissions. However, the high construction and operation costs of IGCC technology are a major challenge for its practical application. Secondly, from 2015 to 2022, keywords such as "Carbon capture and storage" and "CO2 capture" showed strong emergence frequencies, at 501 and 416 respectively, with a relatively long emergence period of 7 years, indicating that these two directions play a crucial role and have objective value for sustainable research. Furthermore, the keyword "CCUS" emerged again in 2021, showing a strong emergence frequency in 2023, and its emergence period continues into 2024, indicating that CCUS technology is a future research trend. Thirdly, from 2023 to 2024, it can be seen that the keywords "Energy transition," "Integrated CO2 capture and utilization," and "Molecular dynamics" will be future research hotspots and trends. Energy transformation is a research direction that combines carbon sequestration with geological energy storage. It utilizes formation pores or mined-out areas as reservoirs, using carbon dioxide as the energy conversion medium. During periods of low electricity demand, compressors use excess electrical energy to compress and store carbon dioxide in the corresponding reservoirs. During peak electricity demand, the high-pressure carbon dioxide is released and converted into electrical energy, thus achieving the goal of energy transformation. Molecular dynamics is used to study the influence of environmental and structural factors on important parameters such as CO2 adsorption, diffusion, interfacial tension, and contact angle. These parameters are crucial for understanding the migration and sequestration behavior of CO2 in geological reservoirs.

[0120] This application also provides an application scenario in which the aforementioned bibliometric analysis method in the CCUS field is applied. Specifically, the bibliometric analysis method in the CCUS field provided in this embodiment can be applied in a bibliometric analysis scenario in the CCUS field. The CCUS field bibliometric analysis scenario includes a literature acquisition stage and a bibliometric analysis stage; core CCUS literature enters the bibliometric analysis stage from the literature acquisition stage, obtaining corresponding overall situation analysis results and co-occurrence analysis results. The bibliometric analysis method in the CCUS field provided in this embodiment belongs to the bibliometric analysis stage. Specifically, in the bibliometric analysis stage targeting CCUS literature, core CCUS literature can be processed to obtain data to be analyzed. VOSviewer software, HistCite software, and Biblioshiny software are used to perform visualization analysis on the data to be analyzed, obtaining a visualization map corresponding to the data to be analyzed. Based on the visualization map, the overall situation analysis results and co-occurrence analysis results of CCUS literature are determined.

[0121] Based on the same inventive concept, this application also provides a CCUS-based bibliometric analysis device for implementing the aforementioned CCUS-based bibliometric analysis method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations of one or more CCUS-based bibliometric analysis device embodiments provided below can be found in the limitations of the CCUS-based bibliometric analysis method described above, and will not be repeated here.

[0122] In one exemplary embodiment, a bibliometric analysis device in the field of CCUS is provided, comprising the following modules.

[0123] The CCUS core literature acquisition module is used to acquire core literature in the CCUS field.

[0124] The core literature processing module is used to process the core literature in the CCUS field to obtain the data to be analyzed.

[0125] The visualization analysis module is used to perform visualization analysis on the data to be analyzed using VOSviewer, HistCite, and Biblioshiny software to obtain the visualization map corresponding to the data to be analyzed.

[0126] The clustering module is used to cluster the data to be analyzed using a trained LDA topic model to obtain topic clustering results.

[0127] The analysis result determination module is used to determine the overall situation analysis results and co-occurrence analysis results of CCUS field literature based on the visualization map and the topic clustering results; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, the analysis results of co-occurring cited literature and the analysis results of co-occurring keywords.

[0128] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0129] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0130] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0131] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0132] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0133] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A bibliometric analysis method in the field of CCUS, characterized in that, The bibliometric analysis methods in the CCUS field include: Obtain core literature in the CCUS field; The core literature in the CCUS field is processed to obtain the data to be analyzed; The data to be analyzed is visualized using VOSviewer, HistCite, and Biblioshiny software to obtain a visual map corresponding to the data, specifically including: A line chart is drawn based on the year and number of articles published in the data to be analyzed. In VOSviewer software, the time range is set from 2000-01-01 to 2024-12-10, the time span is selected as 1 to 4 years, the node types are selected as institutions, countries, co-authors, cited documents and keywords, the link strength is selected as Cosine, the node filtering method is set to g-index of 5, the Pathfinder method is selected, and a knowledge graph of literature-related information network is drawn. In HistCite software, we perform research trend and frontier evolution analysis on the data to be analyzed, obtain research directions and hot topics in the CCUS field, and draw a network knowledge graph based on co-occurrence of keywords. In Biblioshiny software, keyword co-occurrence analysis and cluster analysis were performed on the data to be analyzed to obtain the frequency of keyword occurrence, betweenness centrality, and related research topics, and to draw keyword timeline plots and keyword spurious plots. Line plots, literature-related information network knowledge graphs, keyword co-occurrence-based network knowledge graphs, keyword timeline plots, and keyword spurious plots constitute the visualization map corresponding to the data to be analyzed. The trained LDA topic model is used to cluster the data to be analyzed to obtain topic clustering results; Before determining the overall situation analysis results and co-occurrence analysis results of CCUS literature based on the visualization map and the topic clustering results, the bibliometric analysis method for CCUS literature further includes: optimizing the visualization map based on the g-index and Pathfinder algorithm to obtain an optimized visualization map, specifically including: Construct a composite g-index model; the composite g-index model introduces multi-dimensional parameters or mixed indicators, and the composite methods include: time-weighted composite, discipline normalization, collaborative contribution weighting, and multi-indicator fusion. (1) Time-weighted composite, adjusting the weight of CCUS field literature through a dynamic window: introducing a time decay factor, assigning higher weight to recent citations: ； in, Time-weighted composite weighting for CCUS field literature; The value of g-index represents the threshold size of the collection of high-impact papers by a scholar. For the first i The number of times a paper has been cited; The attenuation rate, This refers to the number of years in which CCUS-related literature has been published. (2) Subject Normalization: The g-index is adjusted based on the average citation level of a subject. The process is as follows: Calculate the average g-index of a subject, which is recorded as the subject baseline value. Divide the individual g-index by the subject baseline value to obtain the normalized composite index. ； in, The normalized composite index; For an individual's g-index; As a benchmark value for the discipline; (3) Collaborative Contribution Weighting: Role Weight Allocation: Differentiating Author Contributions: ； in, Weighted contribution to cooperation; Weighting of author roles; (4) Multi-indicator fusion: Combined with h-index: Construct hg-index to balance quantity and quality: ； in, Indicates hg-index; Indicates h-index; Indicates g-index; and These are the weighting coefficients for h-index and g-index, respectively; The importance of nodes is quantified using a composite g-index model; By integrating local influence, global PageRank, and domain prior weights into a composite g-index model, the final g-index is obtained. The PageRank algorithm, based on the citation network structure of academic papers, measures the importance of a scholar's work within the overall academic network. The formula for calculating the global PageRank is as follows: ； in, For the first The global PageRank of each cited paper; It is the damping factor; for Cited papers; For the thesis Global PageRank; For the thesis The number of outgoing chains; The composite g-index model combines local influence, global PageRank, and domain prior weights, and the calculation formula is as follows: ； in: This is the final g-index; These are local weighting coefficients; For global weight coefficients, satisfying + =1; Score for local influence; This represents the global PageRank score. Prior weights for the domain; The optimized visualization map is obtained based on the final g-index; Based on the visualization map and the topic clustering results, the overall situation analysis results and co-occurrence analysis results of CCUS field literature are determined; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, co-occurring cited literature and co-occurring keywords.

2. The bibliometric analysis method in the field of CCUS according to claim 1, characterized in that, Obtain core literature in the CCUS field, specifically including: Obtain relevant literature in the field of CCUS from both domestic and international sources; Core literature in the field of CCUS was selected from relevant literature both domestically and internationally.

3. The bibliometric analysis method in the field of CCUS according to claim 1, characterized in that, Before using VOSviewer, HistCite, and Biblioshiny software to perform visualization analysis on the data to be analyzed and obtain the corresponding visualization map, the bibliometric analysis method in the CCUS field also includes: The data to be analyzed is deduplicated to obtain the deduplicated data to be analyzed.

4. The bibliometric analysis method in the field of CCUS according to claim 3, characterized in that, The data to be analyzed is deduplicated to obtain deduplicated data to be analyzed, specifically including: Using deduplication software, select the folders named input and output in the CCUS folder, and click Deduplication. The data in the folder named "output" is stored in the folder named "data" to obtain the deduplicated data to be analyzed.

5. A bibliometric analysis device in the CCUS field based on the bibliometric analysis method in the CCUS field as described in claim 1, characterized in that, The bibliometric analysis device in the CCUS field includes: The CCUS core literature acquisition module is used to acquire core literature in the CCUS field. The core literature processing module is used to process the core literature in the CCUS field to obtain the data to be analyzed. The visualization analysis module is used to perform visualization analysis on the data to be analyzed using VOSviewer software, HistCite software, and Biblioshiny software to obtain the visualization map corresponding to the data to be analyzed. The clustering module is used to cluster the data to be analyzed using a trained LDA topic model to obtain topic clustering results; The analysis result determination module is used to determine the overall situation analysis results and co-occurrence analysis results of CCUS field literature based on the visualization map and the topic clustering results; wherein, the overall situation analysis results include the analysis results of the publication time, publishing institution and publishing country of the literature; the co-occurrence analysis results include the analysis results of co-occurring authors, the analysis results of co-occurring cited literature and the analysis results of co-occurring keywords.

6. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the bibliometric analysis method in the field of CCUS as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the bibliometric analysis method in the CCUS field as described in any one of claims 1-4.

Images