本研究將對於臺灣中部地區興建併網型風電系統，對於周遭電力系統進行衝擊預測。基於深度學習技術，提出風速的預測方法，利用歷史數據建立機器學習預測模型，在本研究中，透過實際風電系統長期收集風速風向，與相關環境資料作為訓練資料，使用兩種機器學習方法中常用的遞迴神經網路-長短期記憶網路(Long short-term memory, LSTM)與其變形版本(Gated recurrent unit, GRU)，分別進行短期(未來10分鐘)的風速預測。此外，本研究亦使用即時數位模擬系統(Real Time Digital Simulator, RTDS)建立併網型風電系統模型，以長期風電系統資料與預測結果輸入至模型中產生系統衝擊資料。由結果顯示，在風速預測階段，其中LSTM模型的十分鐘短期預測誤差(Mean absolute percentage error MAPE)為5.6%，三十分鐘短期預測MAPE為8.2%，而GRU模型的十分鐘短期預測MAPE為5.8%，三十分鐘短期預測則為8.5%。基於預測結果的一定準度上，藉由預測風速透過電力電子即時數位模擬器進行更為精確的電力系統衝擊分析，此結果有助於提供臺灣電力公司人員調整調度電力決策，提高電網穩健性。
This study establishes a grid-connected wind power system with a real-time digital simulator (RTDS) model based on a real-world power grid in the central Taiwan to simulate and analyze the impact of the wind power system on the whole power system. Based on deep learning technology, a wind speed prediction method is proposed, and the historical data are used to establish the deep learning prediction model. Long-term wind speed and related environmental information are collected by the wind power system, and additional cloud features are extracted from satellite images to improve the prediction accuracy. With the two common recursive neural network models, the long short-term memory (LSTM) and its modified version (GRU), the wind speed is predicted in a short-term manner (the next 10 minutes) and a medium-term manner (the next 30 minutes). The 10-minute short-term prediction error (MAPE) for the LSTM model is 5.6% and the 30-minute short-term prediction MAPE is 8.2%. In addition, the 10-minute short-term prediction MAPE for the GRU model is 5.8% and the 30-minute short-term prediction MAPE is 8.5%. Based on these prediction accuracy results, a real-time digital simulator is used to examine the impact of the wind speed system on the whole power system. The simulation results help Taiwan Power Company to make proper power dispatch decisions, so that the grid stability can be largely improved.
Table of Contents
致謝 i
摘要 ii
Abstract iii
Table of Contents iv
List of Figures vii
List of Tables xi
Chapter 1 Introduction 1
1.1 Background 1
1.2 Motivation and propose 6
1.3 Contribution of this thesis 8
1.4 Thesis organization 8
Chapter 2 Literature Review 9
2.1 Wind turbine model 9
2.2 Power systems 12
2.2.1 Power system stability 12
2.2.2 Power system simulator 14
2.3 Grid-Connected Wind power systems 16
2.4 Wind speed prediction 19
2.4.1 Artificial Neural Network Model 21
2.4.2 Recurrent Neural Network 22
Chapter 3 Materials and Methods 25
3.1 Framework of the research method 25
3.2 Anemometer 26
3.3 Real-time digital simulator 27
3.3.1 Simulation and data sources 28
3.3.2 Process of RTDS simulation 30
3.4 Deep learning methods for wind speed prediction 32
3.4.1 Long short-term memory 32
3.4.2 Gated recurrent unit (GRU) 34
3.4.3 Time interval for wind speed prediction 36
Chapter 4 Results and Discussion 37
4.1 The system architecture of the RTDS model 37
4.2 Deep learning methods for predicting wind speed 38
4.2.1 Data preprocessing and feature selection. 40
4.2.2 Dataset splitting and hyperparameters search 42
4.2.3 Measurement and prediction performance evaluation 50
4.2.4 Performance of the GRU Model 51
4.3 Case studies on the impact brought by the grid-connected wind power system 53
4.3.1 Case study: Drastically changing wind speed 55
4.3.2 Case study: Sudden shutdown of wind power systems 65
4.3.3 Case study: Long-term analysis 67
4.3.4 Case study: Installed capacity of the grid-connected wind power system 69
4.4 Prediction of Impact for a Grid-Connected wind power system 72
4.4.1 Data Pre-Processing for the Impact Prediction 72
4.4.2 Performance indexes for the impact prediction 76
Chapter 5 Conclusion 84
References 86
經濟部能源局。2019。能源統計月報。網址：hthttps://www.twtpo.org.tw/。上網日期：20019-08-06。
台灣電力公司。2018。台灣電力股份有限公司輸電系統規劃準則。網址：https://www.taipower.com.tw/upload/146/2017111318193843329.pdf。上網日期：20019-08-06。
Bali, V., Kumar, A. and Gangwar, S. 2019. Deep Learning based Wind Speed Forecasting-A Review. 2019 9th International Conference on Cloud Computing, Data Science & Engineering, Noida, India, pp. 426-431
Bae, K. Y., Jang, H. S., & Sung, D. K. 2016. Hourly solar irradiance prediction based on support vector machine and its error analysis. IEEE Transactions on Power Systems 32(2): 935-945.
Bekiroglu, E. and Yazar, M. D. 2019. Analysis of Grid Connected Wind Power System. 2019 8th International Conference on Renewable Energy Research and Applications (ICRERA), Brasov, Romania, pp. 869-873.
Buhan, S., Özkazanç, Y. and Çadırcı, I. 2016. Wind Pattern Recognition and Reference Wind Mast Data Correlations With NWP for Improved Wind-Electric Power Forecasts," in IEEE Transactions on Industrial Informatics, vol. 12, no. 3, pp. 991-1004.
Chang, P. C., Yang, R. Y., Lai C. M. 2015. Potential of Offshore Wind Energy and Extreme Wind Speed Forecasting on the West Coast of Taiwan. Energies. 8. 1685-1700. 10.3390/en8031685.
Energiewende, A. and Sandbag. 2019. The European Power Sector in 2018: Up-to-Date Analysis on the Electricity Transition (Berlin: January 2019), p. 3,
Foley, A. M., Leahy P. G., Marvuglia, A., McKeogh, E. J. 2012. Current methods and advances in forecasting of wind power. (2012) Renewable Energy, 37 (1), pp. 1-8.
Guo, J., Yue, P., Si, J., Chen, Q. and Sun, Y. 2018. Modeling and Simulation of Power Grid Characteristics Caused by Grid-connected Wind Power Generators. 37th Chinese Control Conference (CCC), Wuhan, 2018, pp. 8465-8471.
Haiqiang, Z., Yusheng, X., Jizhu, G. and Jiehui, C. 2017. Ultra-short-term wind speed forecasting method based on spatial and temporal correlation models. The Journal of Engineering, vol. no. 13, pp. 1071-1075.
Hu, Q., Su, P., Yu, D. and Liu, J. 2014. Pattern-Based Wind Speed Prediction Based on Generalized Principal Component Analysis. IEEE Transactions on Sustainable Energy, vol. 5, no. 3, pp. 866-874.
Hu, W., Su, C. and Chen, Z. 2011. Impact of wind shear and tower shadow effects on power system with large scale wind power penetration. IECON 2011 - 37th Annual Conference of the IEEE Industrial Electronics Society, Melbourne, VIC, pp. 878-883.
Linh, N. T. 2009. Voltage stability analysis of grids connected wind generators. 2009 4th IEEE Conference on Industrial Electronics and Applications. Xi'an, pp. 2657-2660.
Lin, Y., Kruger, U., Zhang, J., Wang, Q., Lamont, L. and Chaar, L. E. 2015. Seasonal Analysis and Prediction of Wind Energy Using Random Forests and ARX Model Structures. IEEE Transactions on Control Systems Technology, vol. 23, no. 5, pp. 1994-2002.
Mao, M., Ling, J., Chang, L., Hatziargyriou, N. D., Zhang, J. and Ding, Y. 2016. A Novel Short-Term Wind Speed Prediction Based on MFEC. IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 4, no. 4, pp. 1206-1216.
Mathe, R. M. and Folly, K. A. 2017. Impact of large scale grid-connected wind generators on the power system network. 2017 IEEE PES PowerAfrica, Accra, pp. 328-333.
Pingping, H., Ming, D. and Binbin, L. 2010. Study on transient stability of grid-connected large scale wind power system," The 2nd International Symposium on Power Electronics for Distributed Generation Systems, Hefei, pp. 621-625.
Ren, Y., Suganthan, P. N. and Srikanth, N. 2015. A Comparative Study of Empirical Mode Decomposition-Based Short-Term Wind Speed Forecasting Methods. IEEE Transactions on Sustainable Energy, vol. 6, no. 1, pp. 236-244.
Sharma A., Ola S. R. and Mahela O. P. 2016. Impact of grid disturbances on the output of grid connected wind power generation. 2016 IEEE 1st International Conference on Power Electronics, Intelligent Control and Energy Systems (ICPEICES), Delhi, pp. 1-6.
Sandhu, N. S. and Chanana, S. 2014. Behaviour of wind turbine under different operating modes. 2014 IEEE 6th India International Conference on Power Electronics (IICPE), Kurukshetra, pp. 1-5.
Wang, X., Guoc, P., Huang, X. 2011. A Review of Wind Power Forecasting Models. ICSGCE 2011, 27-30 September 2011.
Yue, C. D., Liu, C. C., Tu, C.C., Lin, T. H. 2019. Prediction of Power Generation by Offshore Wind Farms Using Multiple Data Sources. Energies. 12. 700. 10.3390/en12040700.
被引用:0
點閱:239
評分:
下載:0
書目收藏:0
