.png){kind=icon}
.png){kind=icon}
Return
Research on text information recognition and mining methods for fault records of traction power supply equipment
.png){kind=icon}
OA
AI
L
T
H
Y
Y
W
Z
DOI：10.3389/ffutr.2025.1601538.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The fault records of traction power supply equipment contain rich historical fault processing experience, which is of great significance to the fault handling of traction power supply equipment. However, the fault records of TPSE are unstructured text data, and manual processing of them is time-consuming, labor-intensive, and inefficient. Therefore, the fault records have long been left idle in data systems, lacking exploration and application. In view of this situation, this paper proposes an entity information recognition method for fault records based on the BERT-BiLSTM-CRF algorithm, achieving automated and efficient mining of fault record information. Subsequently, based on the recognized entity information from fault records, a knowledge graph for traction power supply equipment fault handling is constructed. Finally, the retrieval capability of the knowledge graph is improved through an entity similarity-based fast retrieval algorithm, and a decision-making method for fault handling in traction power supply equipment is proposed. This method can quickly associate and recommend similar historical fault handling cases for current equipment faults, thus facilitating knowledge sharing and assisting in enhancing the efficiency and intelligence level of fault handling for maintenance operators.
Keywords:
rail transportation
text mining
knowledge Engineering
expert systems
traction power supplies
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
F
IF:
1.5
Papers:
19
Citations:
194
Organization
No organization information available
Cited Papers