Knowledge Graph-Based In-Context Learning for Advanced Fault Diagnosis in Sensor Networks DOI Creative Commons
Xin Xie, Junbo Wang, Yu Han

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(24), P. 8086 - 8086

Published: Dec. 18, 2024

This paper introduces a novel approach for enhancing fault diagnosis in industrial equipment systems through the application of sensor network-driven knowledge graph-based in-context learning (KG-ICL). By focusing on critical role data detecting and isolating faults, we construct domain-specific graph (DSKG) that encapsulates expert relevant to equipment. Utilizing long-length entity similarity (LES) measure, retrieve information from DSKG. Our method leverages large language models (LLMs) conduct causal analysis textual related faults derived networks, thereby significantly accuracy efficiency diagnosis. details series experiments validate effectiveness KG-ICL accurately diagnosing causes locations systems. leveraging LLMs structured knowledge, our offers robust tool condition monitoring management, improving reliability operations sectors.

Language: Английский

A knowledge graph-aided decision guidance method for product conceptual design DOI
Ru Wang,

Yanshao Sun,

Tao Peng

et al.

Journal of Engineering Design, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 40

Published: July 4, 2024

In the fast-paced world of product design, businesses seek a competitive edge by swiftly addressing user requirements and developing precise solutions. Depending on diverse knowledge, conceptual design makes enterprises invest significantly in knowledge management. Following philosophy decision-based graph-aided decision guidance method is proposed to streamline enhance utilisation design. Firstly, decision-making process modelled using Concept-Decision-Knowledge (CDK) model, yielding meta-model CDK (mCDK). A data-augmented BERT-BiLSTM-CRF model adopted extract information from data sources, forming graph (CDK-KG). case for employing problem-solving configuration generated resolve specific problems when new arise. Validation demonstrated through study launch vehicle first second-stage separation system. The results indicate that can automatically resources construct graph. Furthermore, it provide cases configuration, supporting decision-making.

Language: Английский

Citations

5
Terminological Resources for Biologically Inspired Design and Biomimetics: Evaluation of the Potential for Ontology Reuse DOI Creative Commons
Dilek Yargan, Ludger Jansen

Biomimetics, Journal Year: 2025, Volume and Issue: 10(1), P. 39 - 39

Published: Jan. 9, 2025

Biomimetics aims to learn from living systems develop innovative technical artefacts. As it transcends disciplinary boundaries and needs integrate both biological technological knowledge, a domain ontology for biomimetics would be highly desirable. So far, several terminological resources have been designed support the biomimetic development process. This paper examines nine Biologically Inspired Design biomimetics, including taxonomies, thesauri, ontologies. Their benefits limitations structuring or organising knowledge are evaluated against criteria, availability, clarity, machine readability. Our analysis shows that existing little no potential reuse due inconsistent structure, ambiguous class labels, lack of standardisation, availability. Furthermore, resource adequately represents as all suffer in content representation, reusability, infrastructure. In particular, an adequate supporting is lacking; we discuss desiderata such ontology.

Language: Английский

Citations

0
BICAD bio-inspired design method and BICAD assistant design tool DOI
Alkın Yılmaz Akter, Hüdayim Başak

Journal of Engineering Design, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23

Published: March 4, 2025

Language: Английский

Citations

0
AskNatureGPT: an LLM-driven concept generation method based on bio-inspired design knowledge DOI Creative Commons
Liuqing Chen, Zebin Cai,

Wengteng Cheang

et al.

Journal of Engineering Design, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 35

Published: April 2, 2025

Language: Английский

Citations

0
I-Card: A Generative AI-Supported Intelligent Design Method Card Deck DOI
Liuqing Chen,

Wengteng Cheang,

Zhaojun Jiang

et al.

Published: April 25, 2025

Language: Английский

Citations

0
Advancements in knowledge graphs (KGs) for engineering design DOI
Zuoxu Wang, Xinyu Li, Binyang Song

et al.

Journal of Engineering Design, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 4

Published: May 9, 2025

Language: Английский

Citations

0
An artificial intelligence approach for interpreting creative combinational designs DOI Creative Commons
Liuqing Chen, Shuhong Xiao, Yunnong Chen

et al.

Journal of Engineering Design, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 28

Published: July 11, 2024

Combinational creativity, a form of creativity involving the blending familiar ideas, is pivotal in design innovation. While most research focuses on how combinational achieved through elements, this study computational interpretation, specifically identifying 'base' and 'additive' components that constitute creative design. To achieve goal, authors propose heuristic algorithm integrating computer vision natural language processing technologies, implement multiple approaches based both discriminative generative artificial intelligence architectures. A comprehensive evaluation was conducted dataset created for studying creativity. Among implementations proposed algorithm, effective approach demonstrated high accuracy achieving 87.5% 80% 'additive'. We conduct modular analysis an ablation experiment to assess performance each part our implementations. Additionally, includes error cases bottleneck issues, providing critical insights into limitations challenges inherent interpretation designs.

Language: Английский

Citations

2
A Generative Artificial-Intelligence-Based Workbench to Test New Methodologies in Organisational Health and Safety DOI Creative Commons
Andrea Falegnami, Andrea Tomassi, Giuseppe Corbelli

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(24), P. 11586 - 11586

Published: Dec. 11, 2024

This paper introduces a novel generative artificial intelligence workbench specifically tailored to the field of safety sciences. Utilizing large language models (LLMs), this innovative approach significantly diverges from traditional methods by enabling rapid development, refinement, and preliminary testing new methodologies. Traditional techniques in typically depend on slow, iterative cycles empirical data collection analysis, which can be both time-intensive costly. In contrast, our LLM-based leverages synthetic generation advanced prompt engineering simulate complex scenarios generate diverse, realistic sets demand. capability allows for more flexible accelerated experimentation, enhancing efficiency scalability science research. By detailing an application case, we demonstrate practical implementation advantages framework, such as its ability adapt quickly evolving requirements potential cut down development time resources. The introduction represents paradigm shift methodology offering potent tool that combines theoretical rigor with agility modern AI technologies.

Language: Английский

Citations

1
Knowledge Graph-Based In-Context Learning for Advanced Fault Diagnosis in Sensor Networks DOI Creative Commons
Xin Xie, Junbo Wang, Yu Han

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(24), P. 8086 - 8086

Published: Dec. 18, 2024

This paper introduces a novel approach for enhancing fault diagnosis in industrial equipment systems through the application of sensor network-driven knowledge graph-based in-context learning (KG-ICL). By focusing on critical role data detecting and isolating faults, we construct domain-specific graph (DSKG) that encapsulates expert relevant to equipment. Utilizing long-length entity similarity (LES) measure, retrieve information from DSKG. Our method leverages large language models (LLMs) conduct causal analysis textual related faults derived networks, thereby significantly accuracy efficiency diagnosis. details series experiments validate effectiveness KG-ICL accurately diagnosing causes locations systems. leveraging LLMs structured knowledge, our offers robust tool condition monitoring management, improving reliability operations sectors.

Language: Английский

Citations

0