臺灣博碩士論文加值系統

隨著能源危機與環境汙染日益嚴重，可再生能源如風力發電、水力發電、太陽能發電、生質能發電逐漸受到重視，其中風力發電又為可再生能源發展重點。由於風力發電機非計畫性停機可能造成發電廠的鉅額損失，因此發電廠藉由安裝感測器蒐集發電機與主要零件之數據以及時檢視主要零組件的健康狀態。然而，隨著風力發電機的規模日益增加，所蒐集到的資料也隨之龐大，使得監控人員難以標記每一筆資料。此外當風力發電機更換零組件或設備老化時會產生概念漂移，模型無法正確預測風力發電機發生異常時機而造成預測準確率下降。相關研究已提出許多設備故障偵測方法以解決過高的運行和維護(Operation and Maintenance, O&M)成本問題。其主要分析風力發電機蒐集之多種類型資料，並進行預兆式診斷與健康管理，然而其實驗之資料大多為已標記資料且未考慮概念漂移之情況。
因此本研究提出自我組織映射圖(Self-Organizing Maps, SOM)處理未標記資料(unlabeled data)，將歷史資料依自身的相似性進行分群，並根據分群結果來建立故障偵測超球模型接著，採用兩階層式概念漂移偵測方法判斷風力資料樣本是否發生概念漂移。第一階層是採用「概念漂移偵測法」(Drift Detection Method，DDM)判斷結果是否超過漂移界限，因為DDM隨著時間推移容易模型出現判斷失效，所以第二階層進而採用「置換檢驗」用來確認是否真正發生概念漂移。最後，本研究將透過在不同異常狀況下驗證模型之可行性與使用不同概念的風力資料來測試兩階層式概念漂移偵測方法之能力，當檢測到概念漂移後，採用調適處理來提升模型整體準確度。

Energy crisis and environmental pollution become serious increasingly, so renewable energy, such as wind, solar, hydro, and biomass power, has gradually received attention. The issue of wind power is regarded as a priority to development. Because unplanned downtime may cause a huge loss of wind power plant, they collect data from the generator and main parts by installing sensors and inspect the health status of components. However, as the scale of wind turbines increases, and so does the amount of data collected. It is difficult for monitoring staffs to label data. Furthermore, when replacing components or facility aging may lead to concept drifts, the model cannot precisely predict the abnormalities. Related researches have proposed many methods of machine fault detection to solve excessive operation and maintenance cost. They mainly analyze various types of data collected by wind turbines and conduct prognostics and health management. However, most of the experimental data are labeled and the concept drift is not considered. Thus, this study adopts self-organizing maps (SOM) to process unlabeled data based on their similarity. According to the clustering result, we establish a fault detection hypersphere model. After that, this study adopts a two-stage concept drift detection method to determine whether a concept drift occurs or not. At the first step, the drift detection method (DDM) is applied to determine whether the results exceed the drift bound. However, as time goes by, the DDM is not easy to predict precisely, and hence the second step is to adopt the permutation test to confirm whether the concept drift occurs certainly. Finally, this study will verify the feasibility of the model under different abnormal conditions and employ different concepts of wind dataset to test the capability of a two-step concept drift detection method we proposed. After detecting concept drifts, the overall accuracy will be improved by model adaptation.

摘要 I
ABSTRACT II
致謝 III
目錄 IV
表目錄 V
圖目錄 VI
第一章 緒論 1
第二章 文獻回顧與問題描述 6
2.1 相關文獻探討 6
2.1.1 非監督分群演算法 6
2.1.2 故障預測與診斷方法 10
2.1.3 概念漂移檢測方法 11
2.2 問題描述 12
第三章 研究方法 13
3.1 風力發電機故障偵測模型 13
3.1.1 收集風力發電機未標記資料 13
3.1.2 採用自我組織映射圖將資料做分群 14
3.1.3 建立故障偵測超球模型 16
3.1.4 輸出風力發電機感測點之預測值 18
3.1.5 輸出整體設備健康基準線、健康指數與感測點健康指數 20
3.2 概念漂移偵測 23
3.2.1 第一階層：概念漂移偵測法 23
3.2.2 第二階層：置換檢驗 25
3.3 調適處理 26
第四章 實驗結果 28
4.1. 系統介紹與應用 28
4.2. 資料蒐集 34
4.3. 績效分析 35
4.4. 故障偵測能力 37
4.5. 準確度與精確度分析 38
4.6. 概念漂移實驗分析 41
第五章 結論 44
參考文獻 45

Blaabjerg, F., & Ma, K. (2013). Future on power electronics for wind turbine systems. IEEE Journal of Emerging and Selected Topics in Power Electronics, 1(3), 139-152.
Lu, B., Li, Y., Wu, X., & Yang, Z. (2009). A review of recent advances in wind turbine condition monitoring and fault diagnosis. In 2009 IEEE Power Electronics and Machines in Wind Applications IEEE Press, pp. 1-7.
Milborrow, D. (2006). Operation and maintenance costs compared and revealed. Wind-stats Newsletter, 19, 1-3.
Hatch, C. (2004). Improved wind turbine condition monitoring using acceleration enveloping. Orbit, 61, 58-61.
Márquez, F. P. G., Tobias, A. M., Pérez, J. M. P., & Papaelias, M. (2012). Condition monitoring of wind turbines: Techniques and methods. Renewable Energy, 46, 169-178.
McMillan, D., & Ault, G. W. (2007). Quantification of condition monitoring benefit for offshore wind turbines. Wind Engineering, 31(4), 267-285.
Amirat, Y., Benbouzid, M. E. H., Al-Ahmar, E., Bensaker, B., & Turri, S. (2009). A brief status on condition monitoring and fault diagnosis in wind energy conversion systems. Renewable and Sustainable Energy Reviews, 13(9), 2629-2636.
Lu, Q. F., Cao, Z. T., & Ritchie, E. (2004). Model of stator inter-turn short circuit fault in doubly-fed induction generators for wind turbine. In Proc. of 2004 IEEE 35th Annual Power Electronics Specialists Conference (PESC 04), Vol. 2, pp. 932-937.
Popa, L. M., Jensen, B. B., Ritchie, E., & Boldea, I. (2003). Condition monitoring of wind generators. In Proc. of 38th IAS Annual Meeting. Conference Record of the Industry Applications Conference, Vol. 3, pp. 1839-1846.
Yang, W., Tavner, P. J., Crabtree, C. J., Feng, Y., & Qiu, Y. (2014). Wind turbine condition monitoring: technical and commercial challenges. Wind Energy, 17(5), 673-693.
Mrad, N., Foote, P., Giurgiutiu, V., & Pinsonnault, J. (2013). Condition-based mainte-nance. International Journal of Aerospace Engineering, 2013.
Chen, B., Matthews, P. C., & Tavner, P. J. (2015). Automated on-line fault prognosis for wind turbine pitch systems using supervisory control and data acquisition. IET Renewable Power Generation, 9(5), 503-513.
Touret, T., Changenet, C., Ville, F., Lalmi, M., & Becquerelle, S. (2018). On the use of temperature for online condition monitoring of geared systems–A review. Mechanical Systems and Signal Processing, 101, 197-210.
Zhao, R., Wang, D., Yan, R., Mao, K., Shen, F., & Wang, J. (2018). Machine health monitoring using local feature-based gated recurrent unit networks. IEEE Transactions on Industrial Electronics, 65(2), 1539-1548.
Yang, H. H., Huang, M. L., Lai, C. M., & Jin, J. R. (2018). An approach combining data mining and control charts-based model for fault detection in wind turbines. Renewable Energy, 115, 808-816.
Prisacaru, A., Gromala, P. J., Jeronimo, M. B., Han, B., & Zhang, G. Q. (2017). Prognostics and health monitoring of electronic system: A review. In 2017 18th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE). IEEE, Press, pp. 1-11.
Hu, Y., Baraldi, P., Di Maio, F., & Zio, E. (2017). A Systematic Semi-Supervised Self-adaptable Fault Diagnostics approach in an evolving environment. Mechanical Systems and Signal Processing, 88, 413-427.
Moorthy, C. B., Agrawal, A., & Deshmukh, M. K. (2015). Artificial Intelligence Techniques for Wind Power Prediction: A Case Study. Indian Journal of Science and Technology, 8(25).
Liu, X., Li, M., Qin, S. Y., Ma, X. J., & Wang, W. Z. (2016). A predictive fault diag-nose method of wind turbine based on k-means clustering and neural networks. Journal of Internet Technology, 17(7), 1521-1528.
Khosravi, A., Machado, L., & Nunes, R. O. (2018). Time-series prediction of wind speed using machine learning algorithms: A case study Osorio wind farm, Brazil. Applied Energy, 224, 550-566.
Sheela, K. G., & Deepa, S. N. (2013). Neural network based hybrid computing model for wind speed prediction. Neurocomputing, 122, 425-429.
Leahy, K., Gallagher, C., O’Donovan, P., Bruton, K., & O’Sullivan, D. (2018). A robust prescriptive framework and performance metric for diagnosing and predicting wind turbine faults based on SCADA and alarms data with a case study. Energies, 11(7), 1738.
Wilkinson, M. R., Spinato, F., & Tavner, P. J. (2007). Condition monitoring of generators & other subassemblies in wind turbine drive trains. In Proc. of IEEE International Symposium on Diagnostics for Electric Machines, Power Electronics and Drives (SDEMPED 2007), pp. 388-392).
Wang, Y., & Infield, D. (2013). Supervisory control and data acquisition data-based non-linear state estimation technique for wind turbine gearbox condition monitoring. IET Renewable Power Generation, 7(4), 350-358.
Jahromi, A., Piercy, R., Cress, S., Service, J., & Fan, W. (2009). An approach to power transformer asset management using health index. IEEE Electrical Insulation Magazine, 25(2), 20-34.
Yu, S., Abraham, Z., Wang, H., Shah, M., Wei, Y., & Príncipe, J. C. (2019). Concept Drift Detection and Adaptation with Hierarchical Hypothesis Testing. Journal of the Franklin Institute.
Jain, A. K. (2010). Data clustering: 50 years beyond K-means. Pattern recognition letters, 31(8), 651-666.
Park, H. S., & Jun, C. H. (2009). A simple and fast algorithm for K-medoids clustering. Expert systems with applications, 36(2), 3336-3341.
Nayak, J., Naik, B., & Behera, H. S. (2015). Fuzzy C-means (FCM) clustering algorithm: a decade review from 2000 to 2014. In Computational intelligence in data mining-volume 2 (pp. 133-149). Springer, New Delhi.
Bouguettaya, A., Yu, Q., Liu, X., Zhou, X., & Song, A. (2015). Efficient agglomerative hierarchical clustering. Expert Systems with Applications, 42(5), 2785-2797.
Sasirekha, K., & Baby, P. (2013). Agglomerative hierarchical clustering algorithm-a. International Journal of Scientific and Research Publications, 83, 83.
Murthy, V. S. V. S., Vamsidhar, E., Kumar, J. S., & Rao, P. S. (2010). Content-based image retrieval using Hierarchical and K-means clustering techniques. International Journal of Engineering Science and Technology, 2(3), 209-212.
Gama, J., Medas, P., Castillo, G., & Rodrigues, P. (2004). Learning with drift detection. In Proc. of Brazilian Symposium on Artificial Intelligence, pp. 286-295.
Kohonen, T. (1990). The self-organizing map. Proceedings of the IEEE, 78(9), 1464-1480.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., ... & Bengio, Y. (2014). Generative adversarial nets. In Advances in neural information processing systems (pp. 2672-2680).
Carpenter, G. A., & Grossberg, S. (2010). Adaptive resonance theory (pp. 22-35). Springer US.
Jiang, Y., Chen, K. J., & Zhou, Z. H. (2003). SOM based image segmentation. In International Workshop on Rough Sets, Fuzzy Sets, Data Mining, and Granular-Soft Computing(pp. 640-643). Springer, Berlin, Heidelberg.
Vesanto, J., & Alhoniemi, E. (2000). Clustering of the self-organizing map. IEEE Transactions on neural networks,11(3), 586-600.
Hussin, M. F., Kamel, M. S., & Nagi, M. H. (2004). An efficient two-level SOMART document clustering through dimensionality reduction. In International Conference on Neural Information Processing (pp. 158-165). Springer, Berlin, Heidelberg.
Zhang, Z., Verma, A., & Kusiak, A. (2012). Fault analysis and condition monitoring of the wind turbine gearbox. IEEE Transactions on Energy Conversion, 27(2), 526-535.
Li, Z., Jiang, R., Ma, Z., & Liu, Y. (2015). Fault diagnosis of wind turbine gearbox based on kernel fuzzy c-means clustering. In Proc. of International Conference on Renewable Power Generation (RPG 2015), IET Press.
Gibert, K., Marti-Puig, P., Cusidó, J., & Solé-Casals, J. (2018). Identifying Health Status of Wind Turbines by Using Self Organizing Maps and Interpretation-Oriented Post-Processing Tools. Energies, 11(4), 723.
Yin, S., Wang, G., & Karimi, H. R. (2014). Data-driven design of robust fault detection system for wind turbines. Mechatronics, 24(4), 298-306.
Dao, P. B. (2018). Condition monitoring of wind turbines based on cointegration analysis of gearbox and generator temperature data. Diagnostyka, 19(1), 63-71.
Schlechtingen, M., & Santos, I. F. (2011). Comparative analysis of neural network and regression based condition monitoring approaches for wind turbine fault detection. Mechanical Systems and Signal Processing, 25(5), 1849-1875.
Zhang, Y., Zheng, H., Liu, J., Zhao, J., & Sun, P. (2018). An anomaly identification model for wind turbine state parameters. Journal of Cleaner Production, 195, 1214-1227.
Dong, L., Wang, L., Khahro, S. F., Gao, S., & Liao, X. (2016). Wind power day-ahead prediction with cluster analysis of NWP. Renewable and Sustainable Energy Reviews, 60, 1206-1212.
Wang, S., Schlobach, S., & Klein, M. (2011). Concept drift and how to identify it. Web Semantics: Science, Services and Agents on the World Wide Web, 9(3), 247-265.
Klinkenberg, R., & Joachims, T. (2000). Detecting Concept Drift with Support Vector Machines. In ICML (pp. 487-494).
Gama, J., Medas, P., Castillo, G., & Rodrigues, P. (2004). Learning with drift detection. In Proc. of Brazilian Symposium on Artificial Intelligence, pp. 286-295.
Baena-García, M., del Campo-Ávila, J., Fidalgo, R., Bifet, A., Gavaldà, R., & Mo-rales-Bueno, R. (2006). Early drift detection method. In Proc. of 4th International Workshop on Knowledge Discovery from Data Streams, vol. 6, pp. 77-86.
Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14(1), 35-62.
Huang, K. Y., Chang, W. R., & Yen, H. T. (1990). Self-organizing neural network for picking seismic horizons. In SEG Technical Program Expanded Abstracts 1990, pp. 313-316. Society of Exploration Geophysicists.
Smith, K. A., & Ng, A. (2003). Web page clustering using a self-organizing map of user navigation patterns. Decision Support Systems, 35(2), 245-256.
Guo, P., Infield, D., & Yang, X. (2012). Wind turbine generator condition-monitoring using temperature trend analysis. IEEE Transactions on sustainable energy, 3(1), 124-133.
Good, P. (2013). Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses. Springer Science & Business Media.
Huang, L., Joseph, A. D., Nelson, B., Rubinstein, B. I., & Tygar, J. D. (2011). Adversarial machine learning. In Proceedings of 4th ACM workshop on Security and artificial intelligence, ACM Press, pp. 43-58.