應用混合型神經網路於風力發電機預測狀態監測｜國立臺灣科技大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	王裕傑 Yu-Chieh Wang
論文名稱：	應用混合型神經網路於風力發電機預測狀態監測 Applying Hybrid Neural Networks to Proactive Condition Monitoring of Wind Turbine Generators
指導教授：	張宏展 Hong-Chan Chang
口試委員:	李俊耀 Jun-Yao Li 張建國 Jian-Guo Jhang 郭政謙 Zheng-Qian Guo 張宏展 Hong-Chan Chang
學位類別：	碩士 Master
系所名稱：	電資學院 - 電機工程系 Department of Electrical Engineering
論文出版年：	2023
畢業學年度：	111
語文別：	中文
論文頁數：	111
中文關鍵詞：	風力發電、預測狀態監測、一維卷積神經網路、雙向長短期記憶神經網路、門控循環神經網路
外文關鍵詞：	Wind Power Generation, Predictive State Monitoring, One-dimensional Convolutional Neural Network, Bidirectional Long Short-Term Memory Neural Network, Gated Recurrent Unit Neural Network
相關次數：	點閱：538 下載：8
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

風力發電作為現今不可或缺的永續能源之一，雖然風力發電的技術日益成熟，風力發電機仍需要監測各項指標，以保持長期性的穩定運轉，確保經濟效應，因此風力發電機的運轉維護策略成為了重要的研究議題。現今大多數採用定期性維護保養的方式，雖然可以將故障機率降低，但容易造成資源的浪費，因此本研究提出風力發電機運轉狀態監測方法，搭配線上動態調整警戒門檻，若觸碰到警戒門檻，則立即通知人員處理，達到事先預防發生故障的目的。
本研究提出一套應用混合型神經網路於風力發電機預測狀態監測系統，首先利用量測平台擷取雲林麥寮案場660kW的風力發電機內的振動、電氣、溫度訊號，計算監測指標：發電機驅動端速度、發電機驅動端位移、齒輪箱溫度、電流變化率。其次，利用二種混合型神經網路，分別為一維卷積神經網路(One-Dimensional Convolutional Neural Network)搭配雙向長短期記憶神經網路(Bidirectional Long Short-Term Memory Neural Network)及一維卷積神經網路搭配門控循環神經網路(Gated Recurrent Units Neural Network)，進行日前歷史資料離線擬合訓練及線上預測模組性能評估。模擬結果顯示，在日前歷史資料離線擬合訓練中一維卷積神經網路搭配雙向長短期記憶神經網路有較佳之成效，且發現4項監測指標中資料長度為150分鐘時MAPE值最佳，其中發電機驅動端速度值變化最為激烈，其MAPE值為9.38%。在線上預測案例分析模組性能評估時，4項指標在資料長度為150分鐘時MAPE皆為最低值，且未達警戒標準，是為正常運轉狀態。最後，也對各資料長度筆數之影響線上預測模組上下限門檻值進行探討。

Wind power generation is an indispensable sustainable energy source in today's world. Although the technology of wind turbines is increasingly mature, various indicators still need to be monitored to ensure long-term stable operation and economic efficiency. Therefore, the operational maintenance strategy of wind turbines has become an important research topic. Currently, most wind turbines adopt periodic maintenance, which reduces the probability of failures but can result in resource waste. This study proposes a method for monitoring the operational status of wind turbines by using online dynamic adjustment of alarm thresholds. When the alarm thresholds are exceeded, personnel are immediately notified to take action, thereby achieving the goal of preventing failures in advance.
In this study, a hybrid neural network-based predictive state monitoring system for wind turbines is proposed. Firstly, vibration, electrical, and temperature signals from a 660KW wind turbine at the Mailiao Wind Farm in Yunlin County, Taiwan, are captured using a measurement platform. Monitoring indicators such as generator drive-end speed, generator drive-end displacement, gearbox temperature, and current variation rate are calculated. Secondly, two types of hybrid neural networks are employed: a one-dimensional convolutional neural network (CNN) combined with a bidirectional long short-term memory (LSTM) neural network, and a one-dimensional CNN combined with gated recurrent units (GRU) neural network. These networks are used for offline fitting training of historical data and online prediction module performance evaluation. Simulation results show that the one-dimensional CNN combined with bidirectional LSTM neural network performs better in the offline fitting training of historical data. Among the four monitoring indicators, the data length of 150 minutes yields the best mean absolute percentage error (MAPE) value, with the generator drive-end speed exhibiting the most significant variation, with an MAPE value of 9.38%. In the online prediction module performance evaluation, all four indicators achieve the lowest MAPE value when the data length is 150 minutes, and they do not exceed the alarm threshold, indicating normal operational status. Finally, the study explores the influence of different data length numbers on the upper and lower threshold values of the online prediction module.

中文摘要    I
ABSTRACT    II
誌　　謝    IV
目　　錄    V
圖目錄    VII
表目錄    XIII
第一章    緒　　論……………1
1.1    研究背景與動機……………1
1.2    文獻探討……………4
1.3    研究範疇與流程……………6
1.4    章節概述……………8
第二章    風力發電機運轉狀態監測訊號量測與國際規範……………9
2.1    前言……………9
2.2    案場實驗數據及訊號量測平台介紹……………9
2.3    風力發電機運轉狀態監測指標……………14
2.3.1    齒輪箱溫度……………15
2.3.2    發電機驅動端速度值……………15
2.3.3    發電機驅動端位移值……………20
2.3.4    電流變動率……………21
2.4    風力發電機國際規範之彙整……………22
第三章    基於人工智慧之預測模型……………23
3.1    前言……………23
3.2    基於人工智慧時間序列預測模型之應用……………23
3.2.2    門控循環神經網路……………24
3.2.3    雙向長短期記憶神經網路……………34
3.2.4    一維卷積神經網路……………38
3.2.5    混合型神經網路模型……………43
3.3    30分鐘前預測模組執行流程……………43
第四章    風力發電機預測狀態監測……………47
4.1    前言……………47
4.2    風力發電機預測狀態監測系統與流程……………47
4.2.1    日前歷史資料擬合之訓練……………47
4.2.2    預測30分鐘之線上狀態監測……………51
第五章    實驗案例分析及探討……………53
5.1    實驗案例設計……………53
5.2    實驗案例分析……………54
5.2.1    日前訓練歷史資料擬合評估………………………………54
(1)預測模型對於資料擬合之影響………………………………57
(2)資料長度對於資料擬合之影響………………………………62
5.2.2    線上預測模組性能評估……………75
(1)資料長度對於模組性能及預測未來30分鐘之評估………78
(2)資料長度對於警戒門檻之影響………………………………81
5.3    實驗結果與討論……………85
第六章    結論與未來展望……………87
6.1    結論……………87
6.2    未來展望……………88


                                

[1] T. K. Mahmoud, Z. Y. Dong and J. Ma, "A Developed Integrated Scheme Based Approach for Wind Turbine Intelligent Control,"IEEE Transactions on Sustainable Energy, vol. 8, no. 3, pp. 927-937, July 2017.
[2] Santolamazza, A. Dadi, D. Introna, V. "A Data-Mining Approach for Wind Turbine Fault Detection Based on SCADA Data Analysis Using Artificial Neural Networks,". Energies, vol. 14, pp. 1845, 2021.
[3] L. Yang, L. Wang, Z. Zheng and Z. Zhang, "A Continual Learning-Based Framework for Developing a Single Wind Turbine Cybertwin Adaptively Serving Multiple Modeling Tasks," IEEE Transactions on Industrial Informatics, vol. 18, no. 7, pp. 4912-4921, July 2022.
[4] 張承恩，「域適應網路於感應馬達跨域故障診斷之比較研究」，碩士論文，台灣科技大學電機工程學系，2022年。
[5] H. S. Dhiman, D. Deb, S. M. Muyeen and I. Kamwa, "Wind Turbine Gearbox Anomaly Detection Based on Adaptive Threshold and Twin Support Vector Machines," IEEE Transactions on Energy Conversion, vol. 36, no. 4, pp. 3462-3469, Dec. 2021.
[6] X. Jin, Z. Xu and W. Qiao, "Condition Monitoring of Wind Turbine Generators Using SCADA Data Analysis," IEEE Transactions on Sustainable Energy, vol. 12, no. 1, pp. 202-210, Jan. 2021.
[7] W. Udo and Y. Muhammad, "Data-Driven Predictive Maintenance of Wind Turbine Based on SCADA Data," IEEE Access, vol. 9, pp. 162370-162388, 2021.
[8] A. Heydari, D. A. Garcia, A. Fekih, F. Keynia, L. B. Tjernberg and L. De Santoli, "A Hybrid Intelligent Model for the Condition Monitoring and Diagnostics of Wind Turbines Gearbox," IEEE Access, vol. 9, pp. 89878-89890, 2021.
[9] Y. Peng, W. Qiao, F. Cheng and L. Qu, "Wind Turbine Drivetrain Gearbox Fault Diagnosis Using Information Fusion on Vibration and Current Signals," IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-11, 2021.
[10] X. Jin, Z. Que, Y. Sun, Y. Guo and W. Qiao, "A Data-Driven Approach for Bearing Fault Prognostics," IEEE Transactions on Industry Applications, vol. 55, no. 4, pp. 3394-3401, July-Aug. 2019.
[11] X. Liu, J. Du and Z. -S. Ye, "A Condition Monitoring and Fault Isolation System for Wind Turbine Based on SCADA Data," IEEE Transactions on Industrial Informatics, vol. 18, no. 2, pp. 986-995, Feb. 2022.
[12] M. Fahim, V. Sharma, T. -V. Cao, B. Canberk and T. Q. Duong, "Machine Learning-Based Digital Twin for Predictive Modeling in Wind Turbines," IEEE Access, vol. 10, pp. 14184-14194, 2022.
[13] Morshedizadeh M, "Condition Monitoring of Wind Turbines Using Intelligent Machine Learning Techniques. In: Electronic Theses and Dissertations." 6002. University of Windsor, Ontario, Canada: University of Windsor, pp. 1–96, 2017.
[14] J. Fu, J. Chu, P. Guo and Z. Chen, "Condition Monitoring of Wind Turbine Gearbox Bearing Based on Deep Learning Model," IEEE Access, vol. 7, pp. 57078-57087, 2019.
[15] S. Hong, T. Feng, J. Hu and X. Zhang, "Operation Security Prediction for Wind Turbines Using Convolutional Neural Networks: A Proposed Method," IEEE Systems, Man, and Cybernetics Magazine, vol. 9, no. 1, pp. 4-9, Jan. 2023.
[16] Zheng Li, Xiaorui Luo, Mengjie Liu, Xin Cao, Shenhui Du, Hexu Sun, "Short-Term Prediction of The Power of A New Wind Turbine Based on IAO-LSTM, "Energy Reports, vol. 8, pp. 9025-9037, 2022.
[17] Kou P, Wang C, Liang D, Cheng S, Gao L "Deep Learning Approach for Wind Speed Forecasts at Turbine Locations in A Wind Farm,"IET Renew. Power Gener, vol. 14, pp.2416-2428, 2020.
[18] R. Pandit, D. Astolfi, A.M. Tang, D. Infield "Sequential Data-Driven Long-Term Weather Forecasting Models’ Performance Comparison for Improving Offshore Operation and Maintenance Operations," Energies, vol. 15, no. 19:7233, 2022.
[19] Y. Chen et al., "CNN-BiLSTM Short-Term Wind Power Forecasting Method Based on Feature Selection," IEEE Journal of Radio Frequency Identification, vol. 6, pp. 922-927, 2022.
[20] H.J. Li , J.H. Deng, S. Yuan, P. Feng and Dimuthu D. K. Arachchige. "Monitoring and Identifying Wind Turbine Generator Bearing Faults Using Deep Belief Network and EWMA Control Charts," Front. Energy Res., Sec. Wind Energy , vol. 9, 2021.
[21] Su, X. "Spatial-temporal Attention and GRU Based Interpretable Condition Monitoring of Offshore Wind Turbine Gearboxes," IET Renew. Power Gener, vol.16, pp. 402– 415, 2022.
[22] Jack P. Salameh, Sebastien Cauet, Erik Etien, Anas Sakout, and Laurent Rambault, "Gearbox Condition Monitoring in Wind Turbines: A Review," Mechanical Systems and Signal Processing, vol. 111, pp. 251-264, 2018.
[23] Z.P. Liu and L. Zhang, "A Review of Failure Modes, Condition Monitoring And Fault Diagnosis Methods for Large-Scale Wind Turbine Bearings," Measurement, vol. 149, pp. 1-22, 2020.
[24] Ziqian Kong, Baoping Tang, Lei Deng, Wenyi Liu, Yan Han, "Condition Monitoring of Wind Turbines Based on Spatio-Temporal Fusion of SCADA Data by Convolutional Neural Networks and Gated Recurrent Units," Renewable Energy, vol. 146, pp. 760-768, 2020.
[25] NEMA Std. MG1, "Motors and Generators. ", 2009.
[26] IEC 61400-4, "Wind turbines-Part 4: Design Requirements for Wind Turbine Gearboxes.", pp. 146, 2012.
[27] ISO 10816-1, "Mechanical Vibration- Evaluation of Machine Vibration by Measurements on Non-rotating Parts- Part 1： General Guidelines," pp. 19, 1998.
[28] E. W. K. See-To, X. Wang, K. -Y. Lee, M. -L. Wong and H. -N. Dai, "Deep-Learning-Driven Proactive Maintenance Management of IoT-Empowered Smart Toilet," IEEE Internet of Things Journal, vol. 10, no. 3, pp. 2417-2429, 1 Feb.1, 2023.
[29] W. Jianhong, R. A. Ramirez-Mendoza and W. Yanxiang, "Data Driven Strategy for Linear Parameter Varying Control Design," IEEE Access, vol. 10, pp. 70068-70077, 2022.
[30] A. Halbouni, T. S. Gunawan, M. H. Habaebi, M. Halbouni, M. Kartiwi and R. Ahmad, "Machine Learning and Deep Learning Approaches for CyberSecurity: A Review, " IEEE Access, vol. 10, pp. 19572-19585, 2022.
[31] I. Ullah and Q. H. Mahmoud, "Design and Development of RNN Anomaly Detection Model for IoT Networks," IEEE Access, vol. 10, pp. 62722-62750, 2022.
[32] Alex Sherstinsky, "Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network," Physica D: Nonlinear Phenomena, vol. 404, 2020.
[33] K. Greff, R. K. Srivastava, J. Koutník, B. R. Steunebrink and J. Schmidhuber, "LSTM: A Search Space Odyssey," IEEE Transactions on Neural Networks and Learning Systems, vol. 28, no. 10, pp. 2222-2232, Oct. 2017.
[34] Liu, W., Wang, Q., Zhu, Y. "GRU: Optimization of NPI Performance," J Supercomput 76, pp. 3542–3554, 2020.
[35] Shuai Gao, Yuefei Huang, Shuo Zhang, Jingcheng Han, Guangqian Wang, Meixin Zhang, "Qingsheng Lin,Short-Term Runoff Prediction with GRU and LSTM Networks Without Requiring Time Step Optimization During Sample Generation", Journal of Hydrology, vol. 589, 2020.
[36] C. Ma, G. Dai and J. Zhou, "Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM_BILSTM Method, " IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 6, pp. 5615-5624, June 2022.
[37] Singla, P, Duhan, M& Saroha, S, "An Ensemble Method to Forecast 24-H Ahead Solar Irradiance Using Wavelet Decomposition and BiLSTM Deep Learning Network," Earth Sci Inform 15, pp. 291–306, 2022.
[38] Y. Chen et al., "CNN-BiLSTM Short-Term Wind Power Forecasting Method Based on Feature Selection," IEEE Journal of Radio Frequency Identification, vol. 6, pp. 922-927, 2022.
[39] Pin Zhang, Yi Yang, Zhen-Yu Yin, "BILSTM-Based Soil-Structure Interface Modeling," International Journal of Geomechanics, vol. 21, July 2021.
[40] F. Viel, R. C. Maciel, L. O. Seman, C. A. Zeferino, E. A. Bezerra and V. R. Q. Leithardt, "Hyperspectral Image Classification: An Analysis Employing CNN, LSTM, Transformer, and Attention Mechanism," IEEE Access, vol. 11, pp. 24835-24850, 2023.
[41] Yulin Liu, Ran Li, Binyu Xiong, Shaofeng Zhang, Xinan Zhang, Herbert Iu, Tyrone Fernando, "A Novel Vanadium Redox Flow Battery Modelling Method Using Honey Badger Optimization Assisted CNN-BiLSTM, " Journal of Power Sources, vol. 558, 2023.
[42] N. Ren, Z. Fu, D. Zhou, D. Kong, H. Liu and S. Tian, "Jitter Decomposition Meets Machine Learning: 1D-Convolutional Neural Network Approach," IEEE Communications Letters, vol. 25, no. 6, pp. 1911-1915, June 2021.
[43] Z. Kong, B. Tang, L. Deng, W. Liu, and Y. Han, "Condition Monitoring of Wind Turbines Based on Spatio-Temporal Fusion of SCADA Data by Convolutional Neural Networks and Gated Recurrent Units, " Renewable Energy, vol. 146, pp. 760-768, February 2020.
[44] Hua G, Wang S, Xiao M, Hu S, "Research on The Uplift Pressure Prediction of Concrete Dams Based on The CNN-GRU Model," Water. vol.15, no. 2: 319, 2023.
[45] Y. Li, W. Li, B. Zhang and J. Du, "Federated Adam-Type Algorithm for Distributed Optimization With Lazy Strategy," IEEE Internet of Things Journal, vol. 9, no. 20, pp. 20519-20531, 15 Oct.15, 2022.
[46] Sungil Kim, Heeyoung Kim, "A New Metric of Absolute Percentage Error for Intermittent Demand Forecasts," International Journal of Forecasting, vol. 32, pp.669-679, 2016.
[47] Upma Singh, M. Rizwan, "SCADA System Dataset Exploration and Machine Learning Based Forecast for Wind Turbines," Results in Engineering,Volume vol. 16, 2022.

簡易檢索 / 詳目顯示

相關論文