Research on the Construction of Fault Knowledge Graph for Wind Power Hydraulic Equipment

DOI: https://doi.org/10.62517/jes.202402407

Author(s)

Aichun Du1,2, Xiaoyu Ye1,2, Xincai Chen3, Zhengqiang Zhao1,2, Hong Chen1,2,4,*, Xiaoyun Gong5

Affiliation(s)

1Mechanical and Electrical Department, Hami Polytechnic, Hami, Xinjiang, China 2Hami Wind Power Equipment Intelligent Operation and Maintenance Engineering Technology Research Center, Hami, Xinjiang, China 3Zhengzhou Machinery Research Institute, Zhengzhou, Henan, China 4School of Mechanical and Power Engineering, Zhengzhou University, Zhengzhou, Henan, China 5School of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou, Henan, China *Corresponding Author

Abstract

Hydraulic systems are widely used in wind turbines. However, with the extension of the operation time, the failure rate of the hydraulic system gradually increases, which seriously drags down the operation efficiency of the wind turbine. Therefore, it is urgent to find a fast and accurate fault identification method to identify the abnormality of the wind turbine hydraulic system. The hydraulic system failure analysis documents collected over the past 15 years constitute a data set that has been processed by the BIO annotation technique to make it suitable for analysis. Based on the classic Bert-BILSTM-CRF entity recognition model, an optimized version is developed. Firstly, Bert model is used to collect and extract the features of scattered light; then, these features are fused with the output vector of BILSTM model; finally, the CRF model is used to complete the classification of labels. By embedding the strategy of adversarial learning in the BERT architecture, the robustness of the entity recognition model is successfully enhanced. Subsequently, we will analyze the overall architecture of the obtained triad information and save it in the Neo4j graph database to promote its adaptability. Finally, with the help of Python, we have created a system of fault knowledge mapping. The study ultimately revealed that the optimized model achieved a superior performance of 93 on the F1 score. Up by 1.8 percentage points. The proposed model exhibits an enhancement in performance of roughly 56 percentage points over the Bert-BILSTM-CRF model.

Keywords

Knowledge Graph; Troubleshooting; Entity Extraction; Neo4j

References

[1] Cabrera P, Carta J A, Matos C, et. al. Lessons learned in wind-driven desalination systems in the canary islands: Useful knowledge for other world islands. Desalination, 2024, 583. [2] Leng D, You S, Wang R, et. al. Fatigue damage mitigation of monopile offshore wind turbines utilizing a rotational inertia double tuned mass damper (ridtmd) under realistic wind-wave conditions. Ocean Engineering, 2024, 305. [3] Maldonado-Correa J, Torres-Cabrera J, Martín-Martínez S, et. al. Wind turbine fault detection based on the transformer model using scada data. Engineering Failure Analysis, 2024, 162. [4] Ortega-López V, Faleschini F, Hurtado-Alonso N, et. al. Analysis of raw-crushed wind-turbine blade as an overall concrete addition: Stress–strain and deflection performance effects. Composite Structures, 2024, 340. [5] Peng F Z. Impedance sources (z sources) with inherent fault protection for resilient and fire-free electricity grids. Scientific Reports, 2024, 14. [6] Su H, Cao Q, Li Y. Condition-based opportunistic maintenance strategy for multi-component wind turbines by using stochastic differential equations. Scientific Reports, 2024, 14. [7] Pan S J, Wang J B. Failure analysis and technical transformation of hydraulic transmission system of domestic wind turbine. Inner Mongolia Electric Power Technology, 2011, 29(6):60-62. [8] Huang W T, Wen C B. MISG algorithm based wind turbine hydraulic pitch system fault diagnosis. Electrical Machines and Control Applications, 2018, 45(11):110-115. [9] Hao J., Gao H. N., Wang F., et al. A review of hydraulic system fault diagnosis based on big data analysis. Machine Tools and Hydraulics, 2022, 50(10):166-173. [10] Hu C X, Wen C B. Fault diagnosis of wind turbine hydraulic pitch system based on adaptive observer. Electrical Engineering and Electricity, 2019, 11:5-10. [11] Li X P, Xu J H, Guo Z M, et al. Construction and application of knowledge graph of dispatching automation system. China Electric Power, 2019, 52(02):70-77+157. [12] Han T, Huang H S, Yao L G. Entity extraction for aero-engine fault knowledge graph construction. Combined Machine Tools and Automated Machining Technology, 2021(10):69-73+78. [13] Li X Q, Shi T Y, Li P, et al. Research on text-based knowledge extraction method of high-speed railroad signaling equipment faults. Journal of Railway, 2021, 43(03):92-100. [14] Li X Y, Zhang H, Zhou X H. Chinese clinical named entity recognition with variant neural structures based on BERT methods. Journal of Biomedical Informatics, 2020, 107(prepublished). [15] Ye X Z, Shang L, Dong X Z, et al. Research and application of knowledge graph for fault disposal in distribution networks. Grid Technology, 2022, 46(10):3739-3749. [16] Sun J D, Gu X S, Li Y, et al. Research on Chinese entity relation extraction algorithm based on COAE2016 dataset. Journal of Shandong University (Science Edition), 2017, 52(9):7-12+18. [17] Liu T, Zhang S R, Wang C, et al. BERT-BiLSTM hybrid model for intelligent analysis of water conservancy construction accident text. Journal of Hydropower Generation, 2022, 41(07):1-12. [18] Hochreiter S, Schmidhuber J. Long Short-Term Memory. Neural Computation, 1997, 9(8):1735-1780. [19] Gong Faming, Li Hastiness. Research on massive petroleum domain ontology data storage based on Neo4j. Computer Science, 2018, 45(S1):549-554.