{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,11]],"date-time":"2026-11-11T05:04:03Z","timestamp":1794373443605,"version":"4.9.0"},"reference-count":50,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"China Southern Power Grid Co., Ltd.","award":["036000KC23090003 (GDKJXM20231024)"],"award-info":[{"award-number":["036000KC23090003 (GDKJXM20231024)"]}]},{"award":["036000KC23090003 (GDKJXM20231024)"],"award-info":[{"award-number":["036000KC23090003 (GDKJXM20231024)"]}],"id":[{"id":"https:\/\/ror.org\/03hkh9419","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005015","name":"South China University of Technology","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005015","id-type":"DOI","asserted-by":"crossref"}]},{"id":[{"id":"https:\/\/ror.org\/0530pts50","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Power documentation serves as the core guideline for the safe operation of power systems, and its precise retrieval is crucial for ensuring grid stability and safety. In this context, Retrieval-Augmented Generation (RAG) frameworks emerge as an effective technique by combining LLMs with natural language understanding capabilities and a retrieval-based model with traceability. However, existing Retrieval-Augmented Generation (RAG) frameworks face several main challenges for power-system documents: semantic drift caused by non-standardized industry terminology, increased semantic noise due to fixed-window segmentation, and knowledge conflicts in the multi-source retrieval context. To address these challenges, we propose a multi-path adaptive fusion retrieval framework based on small language models (SLMs). To map queries to standard terminology, our framework first constructs a common terminology repository and section-structure-aware index for the power industry while fully preserving the physical hierarchical logic from related documents. Subsequently, the SLM in our framework assigns prior weights based on query features and retrieved context, which contributes to adaptive fusion of retrieval paths through confidence assessment and consistency verification. With the help of the fusion process, our method effectively filters retrieval noise and resolves knowledge conflicts. Experimental results on real-world power-document datasets covering dispatch, energy storage and emergency response show that our framework achieves an average recall of 91%, outperforming DENSE and BM25 by 21% and 28% respectively. Compared with other methods, it yields the optimal BERTScore F1 (0.7798) and Rouge-1\/2\/L F1 (0.2430, 0.1588, 0.2098) and achieves the best results in the RAGAS framework evaluation, which significantly enhances the rigor and reliability of the question-answering system in the power engineering domain.<\/jats:p>","DOI":"10.3390\/a19050378","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T17:17:42Z","timestamp":1778519862000},"page":"378","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Multi-Route Search and Adaptive Fusion for Power QA with Small Language Model Guidance"],"prefix":"10.3390","volume":"19","author":[{"given":"Zhijun","family":"Shen","sequence":"first","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qian","family":"Guo","sequence":"additional","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lizhou","family":"Jiang","sequence":"additional","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingkang","family":"Huang","sequence":"additional","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenfan","family":"Yu","sequence":"additional","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinlei","family":"Cai","sequence":"additional","affiliation":[{"name":"Power Dispatch Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510335, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hailin","family":"Pang","sequence":"additional","affiliation":[{"name":"School of Electric Power Engineering, South China University of Technology, Guangzhou 510641, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tao","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Electric Power Engineering, South China University of Technology, Guangzhou 510641, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,11]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"110148","DOI":"10.1016\/j.ijepes.2024.110148","article-title":"Real-time power system dispatch scheme using grid expert strategy-based imitation learning","volume":"161","author":"Xu","year":"2024","journal-title":"Int. J. Electr. Power Energy Syst."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"39441","DOI":"10.1038\/s41598-025-22913-3","article-title":"Optimization of A comprehensive dispatching system based on ant colony algorithm and dynamic weight power dispatching strategy","volume":"15","author":"Ming","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"145138","DOI":"10.1016\/j.jclepro.2025.145138","article-title":"Robust load and energy management in smart grids with Prosumer-Integrated distributed energy resources","volume":"496","author":"Habib","year":"2025","journal-title":"J. Clean. Prod."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Li, P., Zhou, M., Lin, X., Zhou, L., and Cai, P. (2024). An Ancillary Decision-Making Method for Hydropower Station Failure Handling Based on Case-Based Reasoning and Knowledge Graph. Processes, 12.","DOI":"10.3390\/pr12122731"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Tang, W., Zhang, Y., Mao, X., Jia, H., Lv, K., Shan, L., and Jiang, T. (2025). Construction and Application of Knowledge Graph for Power Grid New Equipment Start-Up. Energies, 18.","DOI":"10.3390\/en18205471"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"112310","DOI":"10.1016\/j.engappai.2025.112310","article-title":"ElecBench: A large language model benchmark in electric power domain","volume":"162","author":"Zhang","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Gao, M., Li, M., Ji, T., Wang, N., Lin, G., and Wu, Q. (2023). Key technologies of intelligent question-answering system for power system rules and regulations based on improved BERTserini algorithm. Processes, 12.","DOI":"10.20944\/preprints202311.1513.v1"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"100688","DOI":"10.1016\/j.egyai.2026.100688","article-title":"Power system operational reliability evaluation with retrieval-augmented generation enhanced large language model","volume":"24","author":"Cheng","year":"2026","journal-title":"Energy AI"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Chen, J., Lu, G., Pan, Z., Yu, T., Ding, M., and Yang, H. (2022). Research review of the knowledge graph and its application in power system dispatching and operation. Front. Energy Res., 10.","DOI":"10.3389\/fenrg.2022.896836"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Schimanski, T., Ni, J., Kraus, M., Ash, E., and Leippold, M. (2024). Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Association for Computational Linguistics.","DOI":"10.18653\/v1\/2024.acl-long.105"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Gao, F., Li, Y., Sun, W., and Zhang, H. (2023, January 3\u20135). Intelligent Retrieval and Question Answering Technology for Power Grid Security Standards: A Brief Survey. Proceedings of the 2023 IEEE International Symposium on Product Compliance Engineering-Asia (ISPCE-ASIA), Shanghai, China.","DOI":"10.1109\/ISPCE-ASIA60405.2023.10365862"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"8925","DOI":"10.1038\/s41598-025-91940-x","article-title":"A large language model for advanced power dispatch","volume":"15","author":"Cheng","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"773","DOI":"10.1038\/d41586-023-00816-5","article-title":"GPT-4 is here: What scientists think","volume":"615","author":"Sanderson","year":"2023","journal-title":"Nature"},{"key":"ref_14","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Wu, C., Gan, Y., Ge, Y., Lu, Z., Wang, J., Feng, Y., Shan, Y., and Luo, P. (2024). Llama Pro: Progressive Llama with Block Expansion. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Association for Computational Linguistics.","DOI":"10.18653\/v1\/2024.acl-long.352"},{"key":"ref_16","first-page":"9459","article-title":"Retrieval-augmented generation for knowledge-intensive NLP tasks","volume":"33","author":"Lewis","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_17","unstructured":"Guu, K., Lee, K., Tung, Z., Pasupat, P., and Chang, M. (2020, January 13\u201318). Retrieval augmented language model pre-training. Proceedings of the International Conference on Machine Learning, Virtual."},{"key":"ref_18","first-page":"1","article-title":"ATLAS: Few-shot learning with retrieval augmented language models","volume":"24","author":"Izacard","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3571730","article-title":"Survey of hallucination in natural language generation","volume":"55","author":"Ji","year":"2023","journal-title":"ACM Comput. Surv."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Zhang, W., and Zhang, J. (2025). Hallucination Mitigation for Retrieval-Augmented Large Language Models: A Review. Mathematics, 13.","DOI":"10.3390\/math13050856"},{"key":"ref_21","first-page":"10","article-title":"Large Language Models in Power Systems: Enhancing Control and Decision-Making","volume":"1","author":"Bernadic","year":"2025","journal-title":"Int. J. Innov. Solut. Eng."},{"key":"ref_22","unstructured":"Kandpal, N., Deng, H., Roberts, A., Wallace, E., and Raffel, C. (2023). Large language models struggle to learn long-tail knowledge. Proceedings of the 40th International Conference on Machine Learning, PMLR."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Zhao, Q., Wang, R., Cen, Y., Zha, D., Tan, S., Dong, Y., and Tang, J. (2024, January 12\u201316). LongRAG: A Dual-Perspective Retrieval-Augmented Generation Paradigm for Long-Context Question Answering. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, Miami, FL, USA.","DOI":"10.18653\/v1\/2024.emnlp-main.1259"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Wang, F., Wan, X., Sun, R., Chen, J., and Arik, S.O. (2025). Astute rag: Overcoming imperfect retrieval augmentation and knowledge conflicts for large language models. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Association for Computational Linguistics.","DOI":"10.18653\/v1\/2025.acl-long.1476"},{"key":"ref_25","unstructured":"Chan, H.Y., Ho, K.T., Ma, C., Si, Y., Lin, H.L., and Lam, S.L. (2025). Enhancing Retrieval-Augmented Generation for Electric Power Industry Customer Support. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Lee, J., Yoon, S., and Lim, H. (2026). MP-RAG: Multi-Paraphrasing Enhanced Retrieval-Augmented Generation. Proceedings of the 2025 IEEE International Conference on Big Data (BigData), Macau SAR, China, 8\u201311 December 2025, IEEE.","DOI":"10.1109\/BigData66926.2025.11401424"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Xu, Y., Wang, T., Yuan, Y., Huang, Z., Chen, X., Zhang, B., and Wang, Z. (2025). LLM-Enhanced Framework for Building Domain-Specific Lexicon for Urban Power Grid Design. Appl. Sci., 15.","DOI":"10.3390\/app15084134"},{"key":"ref_28","first-page":"3456","article-title":"A Review of Research Methods on Event Knowledge Graph for Power Dispatching","volume":"46","author":"Qi","year":"2024","journal-title":"J. Electron. Inf. Technol."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Zhang, K., Li, L., Xu, X., Xin, R., Xu, X., and Zhang, P. (2024). Intelligent Q&A System for Power Dispatching Based on Retrieval Augmented Generation. Proceedings of the 2024 6th International Conference on Energy, Power and Grid (ICEPG), Guangzhou, China, 27\u201329 September 2024, IEEE.","DOI":"10.1109\/ICEPG63230.2024.10775438"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1016\/j.gloei.2023.08.009","article-title":"Construction and application of knowledge graph for grid dispatch fault handling based on pre-trained model","volume":"6","author":"Ji","year":"2023","journal-title":"Glob. Energy Interconnect."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Choi, S.L., Jain, R., Feng, C., Emami, P., Zhang, H., Hong, J., and Kroposki, B. (2024). Generative AI for Power Grid Operations (No. NREL\/TP-5D00-91176), National Renewable Energy Laboratory (NREL).","DOI":"10.2172\/2477920"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"872","DOI":"10.1109\/JAS.2025.125498","article-title":"Exploring DeepSeek: A survey on advances, applications, challenges and future directions","volume":"12","author":"Deng","year":"2025","journal-title":"IEEE\/CAA J. Autom. Sin."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"163162","DOI":"10.1109\/ACCESS.2025.3610994","article-title":"A Review of Large Language Models for Energy Systems: Applications, Challenges, and Future Prospects","volume":"13","author":"Mirshekali","year":"2025","journal-title":"IEEE Access"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"150951","DOI":"10.1109\/ACCESS.2025.3599922","article-title":"Review of llms applications in electrical power and energy systems","volume":"13","author":"Amjad","year":"2025","journal-title":"IEEE Access"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"1544","DOI":"10.1016\/j.joule.2024.05.009","article-title":"Exploring the capabilities and limitations of large language models in the electric energy sector","volume":"8","author":"Majumder","year":"2024","journal-title":"Joule"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Ni, M., Zhang, J., Fu, C., Wang, J., Ning, X., and Li, S. (2024, January 20\u201322). ChatGrid: Intelligent Knowledge Q&A for Power Dispatching Control Based on Large Language Models and Retrieval-augmented Generation. Proceedings of the 2024 IEEE 7th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), Chongqing, China.","DOI":"10.1109\/ITNEC60942.2024.10733337"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Wang, Y., Xu, T., and Zhu, Y. (2025). SPR-RAG: Semantic Parsing Retriever-Enhanced Question Answering for Power Policy. Algorithms, 18.","DOI":"10.3390\/a18120802"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"3095","DOI":"10.3390\/smartcities7060121","article-title":"A Retrieval-Augmented Generation Approach for Data-Driven Energy Infrastructure Digital Twins","volume":"7","author":"Ieva","year":"2024","journal-title":"Smart Cities"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Alawadhi, S., and Abbas, N. (2025). Optimizing Retrieval-Augmented Generation for Electrical Engineering: A Case Study on ABB Circuit Breakers. Computer Science & Information Technology (CS & IT)\u2013CSCP 2025, AIRCC Publishing Corporation.","DOI":"10.2139\/ssrn.5265681"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Lu, Y., Peng, J., Xu, X., He, Y., Li, T., Wei, J., and Song, H. (2024, January 25\u201327). A Retrieval-Augmented Generation Framework for Electric Power Industry Question Answering. Proceedings of the 2024 2nd International Conference on Electronics, Computers and Communication Technology, Chengdu, China.","DOI":"10.1145\/3705754.3705771"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Ma, Y., Nie, H., Chen, C., Zhang, J., Jiang, J., Wang, B., and Xia, Y. (2025). A Survey of Retrieval-Augmented Generation (RAG) for Large Language Models. Proceedings of the 2025 International Conference on Trustworthy Big Data and Artificial Intelligence (ICTBAI), Tokyo, Japan, 21\u201323 August 2025, IEEE.","DOI":"10.1109\/ICTBAI68361.2025.00008"},{"key":"ref_42","unstructured":"Thakur, N., Reimers, N., R\u00fcckl\u00e9, A., Srivastava, A., and Gurevych, I. (2021). BEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models. Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks 1, Curran Associates."},{"key":"ref_43","unstructured":"Lin, C.Y. (2004). Rouge: A package for automatic evaluation of summaries. Text Summarization Branches Out, Association for Computational Linguistics."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Bhandari, M., Gour, P.N., Ashfaq, A., Liu, P., and Neubig, G. (2020, January 16\u201320). Re-evaluating evaluation in text summarization. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online.","DOI":"10.18653\/v1\/2020.emnlp-main.751"},{"key":"ref_45","unstructured":"Lin, C.-Y., and Och, F.J. (2004, January 2\u20134). Looking for a few good metrics: Rouge and its evaluation. Proceedings of the NTCIR Workshop, Tokyo, Japan."},{"key":"ref_46","unstructured":"Zhang, T., Kishore, V., Wu, F., Weinberger, K.Q., and Artzi, Y. (2020, January 26\u201330). BERTScore: Evaluating Text Generation with BERT. Proceedings of the 8th International Conference on Learning Representations (ICLR 2020), Addis Ababa, Ethiopia. Available online: https:\/\/openreview.net\/forum?id=SkeHuCVFDr."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Es, S., James, J., Anke, L.E., and Schockaert, S. (2024, January 17\u201322). Ragas: Automated evaluation of retrieval augmented generation. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations, St. Julian\u2019s, Malta.","DOI":"10.18653\/v1\/2024.eacl-demo.16"},{"key":"ref_48","first-page":"333","article-title":"The probabilistic relevance framework: BM25 and beyond","volume":"3","author":"Robertson","year":"2009","journal-title":"Found. Trends  Inf. Retr."},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Karpukhin, V., Oguz, B., Min, S., Lewis, P.S., Wu, L., Edunov, S., and Yih, W.T. (2020, January 16\u201320). Dense Passage Retrieval for Open-Domain Question Answering. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online.","DOI":"10.18653\/v1\/2020.emnlp-main.550"},{"key":"ref_50","unstructured":"Han, H., Wang, Y., Shomer, H., Guo, K., Ding, J., Lei, Y., and Tang, J. (2024). Retrieval-augmented generation with graphs (GraphRAG). arXiv."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/5\/378\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T04:18:54Z","timestamp":1778645934000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/5\/378"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,11]]},"references-count":50,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["a19050378"],"URL":"https:\/\/doi.org\/10.3390\/a19050378","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,11]]}}}