XAI 기반 발전설비 고장 기록 데이터 품질 향상 시스템 개발
Development of System for Enhancing the Quality of Power Generation Facilities Failure History Data Based on Explainable AI (XAI)
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Purpose: The deterioration in the quality of failure history data due to differences in interpretation of failures among workers at power plants and the lack of consistency in the way failures are recorded negatively impacts the efficient operation of power plants. The purpose of this study is to propose a system that classifies power generation facilities failures consistently based on the failure history text data created by the workers. Methods: This study utilizes data collected from three coal unloaders operated by Korea Midland Power Co., LTD, from 2012 to 2023. It classifies failures based on the results of Soft Voting, which incorporates the prediction probabilities derived from applying the predict_proba technique to four machine learning models: Random Forest, Logistic Regression, XGBoost, and SVM, along with scores obtained by constructing word dictionaries for each type of failure using LIME, one of the XAI (Explainable Artificial Intelligence) methods. Through this, failure classification system is proposed to improve the quality of power generation facilities failure history data. Results: The results of this study are as follows. When the power generation facilities failure classification system was applied to the failure history data of Continuous Ship Unloader, XGBoost showed the best performance with a Macro_F1 Score of 93%. When the system proposed in this study was applied, there was an increase of up to 0.17 in the Macro_F1 Score for Logistic Regression compared to when the model was applied alone. All four models used in this study, when the system was applied, showed equal or higher values in Accuracy and Macro_F1 Score than the single model alone. Conclusion: This study propose a failure classification system for power generation facilities to improve the quality of failure history data. This will contribute to cost reduction and stability of power generation facilities, as well as further improvement of power plant operation efficiency and stability.
Abstract
Purpose: The deterioration in the quality of failure history data due to differences in interpretation of failures among workers at power plants and the lack of consistency in the way failures are recorded negatively impacts the efficient operation of power plants. The purpose of this study is to propose a system that classifies power generation facilities failures consistently based on the failure history text data created by the workers. Methods: This study utilizes data collected from three coal unloaders operated by Korea Midland Power Co., LTD, from 2012 to 2023. It classifies failures based on the results of Soft Voting, which incorporates the prediction probabilities derived from applying the predict_proba technique to four machine learning models: Random Forest, Logistic Regression, XGBoost, and SVM, along with scores obtained by constructing word dictionaries for each type of failure using LIME, one of the XAI (Explainable Artificial Intelligence) methods. Through this, failure classification system is proposed to improve the quality of power generation facilities failure history data. Results: The results of this study are as follows. When the power generation facilities failure classification system was applied to the failure history data of Continuous Ship Unloader, XGBoost showed the best performance with a Macro_F1 Score of 93%. When the system proposed in this study was applied, there was an increase of up to 0.17 in the Macro_F1 Score for Logistic Regression compared to when the model was applied alone. All four models used in this study, when the system was applied, showed equal or higher values in Accuracy and Macro_F1 Score than the single model alone. Conclusion: This study propose a failure classification system for power generation facilities to improve the quality of failure history data. This will contribute to cost reduction and stability of power generation facilities, as well as further improvement of power plant operation efficiency and stability.
참고문헌 (30)
- Batelić J, "Impact of Remediation-Based Maintenance on the Reliability of a Coal-Fired Power Plant Using Generalized Stochastic Petri Nets", Energies, 14(18), pp.5682~ (2021)
- Blanco, M. A., "A text-mining approach to assess the failure condition of wind turbines using maintenance service history", Energies, 12(10), pp.1982~ (2019)
- Chawla, N. V., "SMOTE : synthetic minority over-sampling technique", Journal of Artificial Intelligence Research, 16, pp.321~357 (2002)
- Chen, M., "XGBoost-based algorithm interpretation and application on post-fault transient stability status prediction of power system", IEEE Access, 7, pp.13149~13158 (2019)
- Han, M., "The Effect of Domain Specificity on the Performance of Domain-Specific Pre-Trained Language Models", Journal of Intelligence and Information Systems, 28(4), pp.251~273 (2022)
- He, H., "Learning from imbalanced data", IEEE Transactions on Knowledge and Data Engineering, 21(9), pp.1263~1284 (2009)
- Hong, J., "Defect Prediction and Variable Impact Analysis in CNC Machining Process", Journal of the Korean society for Quality Management, 52(2), pp.185~199 (2024)
- Hong, S., "Automated text classification of maintenance data of higher education buildings using text mining and machine learning techniques", Journal of Architectural Engineering, 28(1), pp.04021045~ (2022)
- Hwang, S., "Study of Reliability Analysis Based Power Generation Facilities Maintenance System-Focused on Continuous Ship Unloader-", Journal of the Korean society for Quality Management, 51(2), pp.315~327 (2023)
- Jardine, A. K., "A review on machinery diagnostics and prognostics implementing condition-based maintenance", Mechanical Systems and Signal Processing, 20(7), pp.1483~1510 (2006)
- Kim, K., "A Study on Fault Diagnosis of Boiler Tube Leakage based on Neural Network using Data Mining Technique in the Thermal Power Plant", Transactions of The Korean Institute of Electrical Engineers, 66(10), pp.1445~1453 (2017)
- Kim, M., "A Study on Fault Detection Method Using Weight-based Local Outlier Factor", Journal of Korean Institute of Intelligent Systems, 31(4), pp.273~278 (2021)
- Korea Energy Economics Institute, "KEEI Yearbook of Energy Statistics 2022" (2022), Korea Energy Economics Institute
- Lee, H., "Identifying process capability index for electricity distribution system through thermal image analysis", Journal of Korean Society for Quality Management, 49(3), pp.327~340 (2021)
- Lee, S., "Industrial and Technological Trends of Personal Mobility in Korea and Abroad", Journal of the Korean Society of Automotive Engineers, 37(1), pp.36~45 (2015)
- Lee, S., "A Pre-processing Process Using TadGAN-based Time-series Anomaly Detection", Journal of the Korean society for Quality Management, 50(3), pp.459~471 (2022)
- Lee, Y., "Interpretation of load forecasting using explainable artificial intelligence techniques", The Transactions of The Korean Institute of Electrical Engineers, 69(3), pp.480~485 (2020)
- Lim, C., "A study on development of power grid fault prediction system based on big data and preceding activities to calculate optimal investment cost", Journal of the Korean Data And Information Science Society, 29(3), pp.779~794 (2018)
- Liu, J., "Short text classification for faults information of secondary equipment based on convolutional neural networks", Energies, 15(7), pp.2400~ (2022)
- Machlev, R., "Explainable Artificial Intelligence(XAI)techniques for energy and power systems : Review, challenges and opportunities", Energy and AI, 9, pp.100169~ (2022)
- Meng, Q., "Electric power audit text classification with multi-grained pre-trained language model", IEEE Access, 11, pp.13510~13518 (2023)
- Park, J., "A reliable intelligent diagnostic assistant for nuclear power plants using explainable artificial intelligence of GRU-AE, LightGBM and SHAP", Nuclear Engineering and Technology, 54(4), pp.1271~1287 (2022)
- Rahman, U., "Application of classification models on maintenance records through text mining approach in industrial environment", Journal of Quality in Maintenance Engineering, 29(1), pp.203~219 (2023)
- Ribeiro, M. T., "“Why should I trust you?” Explaining the predictions of any classifier", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.1135~1144 (2016)
- Sarker, I. H, "Deep learning : a comprehensive overview on techniques, taxonomy, applications and research directions", SN Computer Science, 2(6), pp.420~ (2021)
- Tian, X., "A novel classification model SA-MPCNN for power equipment defect text", Transactions on Asian and Low-Resource Language Information Processing, 20(6), pp.1~21 (2021)
- Vong, C. M., "Accurate and efficient sequential ensemble learning for highly imbalanced multi-class data", Neural Networks, 128, pp.268~278 (2020)
- Wang, H., "Short text mining framework with specific design for operation and maintenance of power equipment", CSEE Journal of Power and Energy Systems, 7(6), pp.1267~1277 (2020)
- Zednik, C, "Solving the black box problem : A normative framework for explainable artificial intelligence", Philosophy & Technology, 34(2), pp.265~288 (2021)