Research Article
A domain knowledge graph-driven joint recommendation system for wiring mode and automation configuration in differentiated power supply zones
@ARTICLE{10.4108/ew.14666, author={Min Li and Junhua Gong and Jing Tan and Feipeng Huang and Xiaohong Tan}, title={A domain knowledge graph-driven joint recommendation system for wiring mode and automation configuration in differentiated power supply zones}, journal={EAI Endorsed Transactions on Energy Web}, volume={13}, number={1}, publisher={EAI}, journal_a={EW}, year={2026}, month={10}, keywords={domain knowledge graph, graph neural network, multi-task joint learning, distribution network planning, knowledge reasoning}, doi={10.4108/ew.14666} }- Min Li
Junhua Gong
Jing Tan
Feipeng Huang
Xiaohong Tan
Year: 2026
A domain knowledge graph-driven joint recommendation system for wiring mode and automation configuration in differentiated power supply zones
EW
EAI
DOI: 10.4108/ew.14666
Abstract
INTRODUCTION: Distribution network planning requires jointly determining wiring modes and automation configurations, yet existing methods treat them independently without regulatory constraints. OBJECTIVES: To develop a knowledge graph-driven framework for joint wiring mode and automation configuration recommendation across differentiated supply zones. METHODS: A BERT-CRF pipeline constructs a knowledge graph from regulatory texts. TransR-initialized graph attention networks and CRF decoders with SWRL constraints are coupled via multi-task learning, evaluated on 1,443 records and 12 field tasks. RESULTS: Acc@3 of 0.817, F1 of 0.757, and CCR of 0.733, surpassing the best baseline by 4.3% and 10.4% respectively. Expert validation scores 4.50/5.00 with 73.3% planning time reduction. CONCLUSION: Integrating knowledge graph embedding, GNN aggregation, and constraint-aware joint optimization yields accurate, compliant, and traceable distribution planning recommendations.
Copyright © 2026 Min Li et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.