澎湖電力系統主要能源為柴油發電，在發電成本較高的情況下，需增設新的風力再生能源設備，來降低柴油系統發電成本。因為澎湖夏季風速較低，導至風電效益不高，在不使用海底電纜併聯台灣電力系統的情況下，可以考慮增加風力發電機容量來提高再生能源輸出；但在冬季時風速較高，風能充裕的情況下，過多的再生能源容量將導致風力發電機機輸出太多，使實功率充斥系統導致系統不穩定。
本文利用MATLAB/Simulink建立澎湖電力系統併聯離岸式風力發電系統，並為了使風力發電系統皆可操作在最大功率點及系統實功率達到一個快速又穩定的響應，提出兩種智慧型控制器，包含廣義迴歸類神經網路與粒子群優化-徑向基底類神經網路，將其應用風力發電系統的轉速與旋角控制。感應發電機轉速資訊是經由提出的模型參考適應系統來估測，達到無轉速感測器的實現。在風速較低時，廣義迴歸類神經網路控制器控制發電機轉速，使風力發電系統達成風力最大功率追蹤，提高發電效益。粒子群優化-徑向基底類神經網路控制器控制旋角角度，在冬季風速過高時能夠控制旋角角度降低進風量，減少葉扇損壞機率並使風力發電系統輸出穩定。

The main generation power in Penghu power system is diesel generation system, which generation cost is very high. Therefore, building new wind renewable energy equipments is needed to reduce the generation cost of diesel. Due to the lower wind speed in summer in Penghu, the efficiency of the wind power generation is lower and which can be solved by considering increasing the capacity of wind power equipments to raise the output of renewable energy without using the submarine cables connect to Taiwan. However, in the higher wind speed condition in winter in Penghu, the output of wind power generation system will become too high due to the large capacity of wind power equipments to influence the safety of Penghu power system.
In this thesis, a simulation model for Penghu power generation system with the offshore wind power generation system has developed using MATLAB/Simulink. To achieve a fast and stable response for the real power control, an intelligent controller was proposed, which consists of a General Regress Neural Network (GRNN) and a Particle Swarm Optimization Radial Basis Function Neural Network (PSO-RBFNN) to control the rotor speed and pitch angle. To achieve a sensor-less vector-control strategy for an induction generator (IG), an estimation of the rotor speed is based on model reference adaptive system (MRAS) speed observer theory. In the low wind speed condition, the rotor speed of generator is controlled by GRNN for maximum power point tracking (MPPT). In the excessive wind speed condition, the pitch angle of wind turbine is controlled by PSO-RBF to ensure the stability of wind power generation system.

摘要 i
Abstract ii
目錄 iii
圖次 vi
表次 viii
第一章 緒論 1
1.1 研究背景與動機 1
1.2 文獻回顧 2
1.3 論文貢獻 3
1.4 論文架構 3
第二章 澎湖電力系統簡介 5
2.1 澎湖發電系統介紹 5
2.1.1 尖山火力發電廠 5
2.1.2 中屯風力發電廠 6
2.1.3 湖西風力發電廠 6
2.1.4 輸電線路與電壓器參數 7
2.1.5 負載成長評估 8
2.2 離岸式風場評估 9
2.2.1 離岸式風力機機型與容量 9
2.2.2 離岸式風力占比與規劃 10
第三章 風力發電系統與智慧型控制器理論與應用 13
3.1 風力發電系統理論 13
3.1.1 風能簡介 13
3.1.2 風力發電機原理 13
3.1.3 風力發電系統控制原理 16
3.1.4 感應發電機之數學模型分析 17
3.2 智慧型控制器之理論與應用 22
3.2.1 類神經網路簡介 22
3.2.2 類神經網路控制架構 23
3.2.3 徑向基底類神經網路之簡介與原理 25
3.2.4 廣義迴歸類神經網路之簡介與原理 26
3.3 粒子群最佳化演算法簡介與原理 28
3.3.1 傳統粒子群演算法 29
3.3.2 具時變加速度係數最佳化粒子群演算法 31
第四章 智慧型轉速控制器於最大功率追蹤控制 34
4.1 前言 34
4.2 智慧型轉速控制器之設計理論 34
4.2.1 整體架構 34
4.2.2 基於模型參考適應系統的轉速感測器 35
4.2.3 廣義迴歸類神經網路控制器於轉速控制的設計 41
4.2.4 廣義迴歸類神經網路設計 43
4.2.5 廣義迴歸類神經網路的平滑參數設計 44
第五章 智慧型旋角控制器於風機輸出功率控制 46
5.1 前言 46
5.2 風力發電系統的最大功率追蹤設計原理 46
5.2.1 旋角控制的設計原理 46
5.2.2 比例-積分的旋角控制 46
5.2.3 徑向基底類神經網路-粒子群優化控制器於旋角控制之設計 48
5.2.4 徑向基底類神經網路設計 49
5.2.5 PSO應用於線上學習訓練機制 50
第六章 澎湖發電系統之模擬結果 53
6.1 前言 53
6.2 智慧型控制器模擬結果 53
6.2.1 智慧型轉速控制器模擬結果 53
6.2.2 智慧型旋角控制器模擬結果 59
6.3 澎湖發電系統架構 61
6.4 夏季澎湖發電系統併聯離岸式風場之模擬結果 63
6.4.1 原始澎湖發電系統 63
6.4.2 併聯加入智慧型控制器之離岸式風場的澎湖發電系統 64
6.5 冬季澎湖發電系統併聯離岸式風場之模擬結果 66
6.5.1 原始澎湖發電系統 66
6.5.2 併聯加入智慧型控制器之離岸式風場的澎湖發電系統 68
第七章 結論與未來研究方向 71
7.1 結論 71
7.2 未來研究方向 71
參考文獻 73

[1] S. R. Bull, “Renewable Energy Today and Tomorrow,” Proceedings of the IEEE, vol. 89, Issue 8, pp. 1216-1226, 2001.
[2] 行政院原子能委員會，「風力發電廠區風速及發電量統計預報系統開發」，2010年12月。
[3] M. S. Miranda, S. R. Silva, and R. O. C. Lyra, “An Alternative Isolated Wind Electric Pumping System Using Induction Machines,” IEEE Transactions on Energy Conversion, vol. 14, pp. 1611-1616, Dec. 1999.
[4] A. Grauers, “Efficiency of Three Wind Energy Generator Systems,” IEEE Transactions on Energy Conversion, vol. 11, no. 3, pp. 650-657, 1996.
[5] 周鵬程，「類神經網路入門」，全華科技圖書股份有限公司，2006年1月。
[6] 葉怡成，「類神經網路模式應用與實作」，儒林圖書有限公司，2004年9月。
[7] 張斐章、張麗秋、黃浩倫，「類神經網路-理論與實務」，東華書局，2004年3月。
[8] 張智凱，「應用智慧型最大功率追蹤控制於再生能源發電系統」，國立中山大學電機工程研究所碩士論文，2012年6月。
[9] A. A. Esmin, G. L. Torres, and C. Z. Souza, “A Hybrid Particle Swarm Optimization Applied to Loss Power Minimization,” IEEE Transaction on Power Systems, vol. 20, no. 2, pp. 859-866, 2005.
[10] T. Li, C. Wei, and W. Pei, “PSO with Sharing for Multimodal Function Optimization,” IEEE Int. Conf. Neural Networks and Signal Process., vol. 1, pp. 450-453, 2003.
[11] W. Xueguang, W. Weisheng, D. Huizhu, and C. Yunping, “Application of Models of The Wind Energy Conversion System to Wind Power Dynamic Analysis,” IEEE Power System Technology, vol. 2, pp. 1406-1411, Aug 1998.
[12] C. Schauder, “Adaptive Speed Identification for Vector Control of Induction Motors without Rotational Transducers,” IEEE Trans. Ind. Appl., vol. 28, no. 5, pp. 1054-1061, 1992.
[13] H. Tajima and Y. Hori, “Speed Sensor-less Field Orientation Control of The Induction Machine,” IEEE Trans. Ind. Appl., vol. 29, no. 1, pp. 175-180, 1993.
[14] G. Yang and T. H. Chin, “Adaptive-speed Identification Scheme for A Vector-Controlled Speed Sensor-less Inverter-induction Motor Drive,” IEEE Trans. Ind. Appl., vol. 29, no. 4, pp. 820-825, 1993.
[15] F. Z. Peng and T. Fufao, “Robust Speed Identification for Speed-Sensor-less Vector Control of Induction Motors,” IEEE Trans. Ind. Appl., vol. 30, no. 5, pp. 1234-1240, 1994.
[16] Y. D. Landau, “Adaptive Control – The Model Reference Approach,” Marcel Dekker, New York, 1979.
[17] F. J. Lin, R. J. Wai, and P. C. Lin, “Robust Speed Sensor-less Induction Motor Drive,” IEEE Trans. Aerosp. Electron. Systems, vol. 35, no. 2, pp. 532-542, 1998.
[18] R. Cardenas, R. Pena, J. Proboste, G. Asher, and J. Clare, “MRAS Observer for Sensor-less Control of Standalone Doubly Fed Induction Generators,” IEEE Transactions on Energy Conversion, vol. 20, no. 4, pp. 710-718, Dec 2005.
[19] T. Senjyu, S. Tamaki, N. Urasaki, and K. Uezato, H. Higa, T. Funabashi, H. Fujita, and H. Sekine, “Wind Velocity and Rotor Position Sensor-less Maximum Power Point Tracking Control for Wind Generation System,” 35th Annual IEEE Power Electron. Specialists Conf., pp. 2023-2028, 2004.
[20] 洪志明，「智慧型無轉速感測器之風力發電系統最大功率追蹤控制」，國立中山大學電機工程研究所博士論文，2011年8月。
[21] T. Senjyu, S. Tamaki, N. Urasaki, and K. Uezato, H. Higa, T. Funabashi, H. Fujita, and H. Sekine, “Wind Velocity and Rotor Position Sensor-less Maximum Power Point Tracking Control for Wind Generation System,” 35th Annual IEEE Power Electron. Specialists Conf., pp. 2023-2028, 2004.
[22] R. Cardenas, R. Pena, J. Proboste, G. Asher, and J. Clare, “MRAS Observer for Sensor-less Control of Standalone Doubly Fed Induction Generators,” IEEE Transactions on Energy Conversion, vol. 20, no. 4, pp. 710-718, Dec. 2005.
[23] M. A. Yurdusev, R. Ata, and N. S. Cetin, “Assessment of Optimum Tip Speed Ratio in Wind Turbines Using Artificial Neural Networks,” Energy, vo1. 31, no. 12, pp. 2153-2161, 2006.
[24] Hui Li, K. L. Shi, and P. G. McLaren, “Neural-network-based Sensor-less Maximum Wind Energy Capture with Compensated Power Coefficient,” IEEE Trans. Ind. Appl., vol. 41, no. 6, pp. 1548-1556, 2005.
[25] M. Karrari, W. Rosehart, and O. P. Malik, “Comprehensive Control Strategy for A Variable Speed Cage Machine Wind Generation Unit,” IEEE Transactions on Energy Conversion, vo1. 2, no. 2, pp. 415-423, 2005.
[26] Q. Wang and L. Chang, “An Intelligent Maximum Power Extraction Algorithm for Inverter-based Variable Speed Wind Turbine Systems,” IEEE Transactions on Energy Conversion, vol. 19, no. 5, pp. 1242-1249, 2004.
[27] E. Muljadi and C. P. Butterfield, “Pitch-controlled Variable-speed Wind Turbine Generation,” IEEE Trans. Ind. Appl. vol. 37, no. 1, pp. 240-246, 2001.
[28] R. Sakamoto, T. Senjyu, R. Sakamoto, T. Kaneko, N. Urasaki, T. Takagi, S. Sugimoto, and H. Sekine, “Output Power Leveling of Wind Turbine Generator by Pitch Angle Control Using Control,” in Proc. IEEE PSCE Conf., pp. 2044-2049, 2006.
[29] T. Senjyu, R. Sakamoto, N. Urasaki, T. Funabashi, and H. Sekine, “Output Power Leveling of Wind Farm Using Pitch Angle Control with Fuzzy Neural Network,” in Proc. IEEE Power Engineering Sociey General Metting, June 2006.
[30] J. S. R. Jang and C. T. Sun, “Function Equivalence between Radial Basis Function Networks and Fuzzy Inference Systems,” IEEE Transaction on Neural Networks, vol. 4, no. 1, pp. 156-159, 1993.
[31] S. Seshagiri and H. K. Khail, “Output Feedback Control of Nonlinear Systems Using RBF Function Equivalence between Radial Basis Function Neural Networks,” IEEE Transaction on Neural Networks, vol.11, no. 1, pp. 69-79, 2000.
[32] T. Li, C. Wei and W. Pei, “PSO with Sharing for Multimodal Function Optimization,” IEEE Int. Conf. Neural Networks and Signal Process., vol. 1, pp. 450-453, 2003.