臺灣博碩士論文加值系統

狀態估計是系統工程中的一項重要技術，根據外部量測到的數據估算動態系統中內部狀態，在系統建模上具有重要意涵，在控制工程中亦是回饋控制的一個要角。一般，系統的輸入和輸出進行量測而得到的數據只能反映系統的外部特性，而系統的動態行為就需要用內部狀態變量來描述，傳統之估計方法如卡爾曼濾波器，在動態訊號的估計上有非常優秀之表現，樂為學者經常引用。然從一全域-局部(Global-Local)的權衡中思考，常是因刻意追求局部細膩之精度(Accuracy)，而誤失了其全域之系統動態規律。本研究嘗試在一個新近發展的估計技術「移動區間狀態估測」(MHE)上，引用混合架構，加上一個支持向量機迴歸(SVR)，建立一個SVR-MHE的動態系統估計方法論，藉支持向量機迴歸所建立全域性粗略模型，再令一個動態估計跟隨以為精細估計，本研究使用移動區間之狀態估測結合全域建模的混合模型，主要是因為其具有以下四點優勢。因此建立的模型具有:
1.蘊涵模型參考下的動態系統跟隨估計能力
2.較卡爾曼濾波器更細膩的估計能力
3.較高的全域-局部權衡，可為大尺度規律系統之建模跟隨，作為Model Based控制方法(如LQR)之基石，改善效能
4.因引用支持向量機迴歸而涵括有大邊限抵禦雜訊的能力
本論文利用美國風電場的風速作及台灣風速做為預測範例，再以實驗佐證。並以全域建模的角度詮釋及重新建立一其長時距之系統模型，再施以局部的狀態估計短時距之即時風速預測。實驗結果顯示所提議模型系統具有一定的準確性，進而達到預測該有之成效。

For the topic of state estimation, estimating internal state of a dynamic system has important implications on system modeling, also a part of feedback control in control engineering. In general, the system input and output can only reflect the external features, but the dynamical system behavior needs to be described with the internal state variables. Dynamic signal has very good performance in estimation and often cited by scholars, such as traditional estimation method Klaman filter. However, thinking from a global-local trade-off, often due to the deliberate pursuit of the accuracy of the local, and lost its global system dynamic trend. This study attempted to estimate in newly developing technology ”Moving horizon estimation”, and establish a hybrid architecture with a support vector regression then build a SVR-MHE dynamic system estimation method. With the establishment of a global rough model by SVR, and then make a local dynamic estimate tracking as precise estimation. This study uses MHE to replace the Kalman filter because mainly it has four advantages. First, the ability of dynamic system tracking and estimation under implication model reference. Second, more delicate estimation ability than the Kalman filter. Third, the higher the global-local trade-off for the large-scale regularity of the system of the modeling to track and as the model based control method to improve performance. Fourth, because using SVR and encompassed a large margin against noise ability. Actually, the hybrid model is a key to dynamic system. This study will use the wind speed prediction of wind farms as an example as an experiment. Finally, a system model of longtime-distance is interpreted and re-established from the perspective of global modeling, and then applied local state estimation to estimate instantaneous wind speed at short time intervals. The simulation results of verification show that the proposed system has feasibility, and then achieves the predicted results.

誌 謝 I
目錄 IV
圖目錄 VII
表目錄 IX
第一章　緒論 1
1-1 　前言 1
1-2 　研究動機與目的 4
1-3 　文獻探討 5
1-4 　本論文架構 9
第二章　研究原理與基礎 11
2-1 　建模方法 11
2-1-1 　自迴歸 12
2-1-1 　支向機迴歸 14
2-2 　狀態估測 17
2-2-1 　卡爾曼濾波器 18
2-3 　移動區間狀態估測 22
2-3-1 　最優估計 23
2-3-2 　全訊息估測 25
2-3-3 　移動區間估測 26
2-3-4 　到達代價 28
2-4 　狀態方程式 30
第三章　研究方法及架構模型 33
3-1 　研究流程 33
3-2 　研究設備 34
3-3 　系統模型之評估 35
3-4 　自迴歸與支向機迴歸的狀態方程建立 37
第四章　以風電場之風速預測為例 41
4-1 　短期預測v.s.長期預測 41
4-2 　系統模型之評估 42
4-3 　混合模型架構 43
第五章　實驗與討論 47
5-1 　全域建模方法比較 47
5-2 　局部估測方方法比較 51
5-3 　自迴歸建模與支向機建模驗證 53
5-4 　局部追隨估測演算法之驗證 56
5-5 　混合模型性能之評估 58
第六章　結論 72
6-1 　結論 72
6-2 　未來計畫與發展 73
參考文獻 74

[1]B. W. Bequette, “Nonlinear Predictive Control Using Multi-rate Sampling,” Chemical Engineering, vol. 69, no. 1, pp. 136–143, February 1991.
[2]Y. Ramamurthi, P. B. Sistu, and B. W. Bequette, “Control-relevant dynamic data reconciliation and parameter estimation,” Computers and Chemical Engineering, vol. 17, no. 1, pp. 41–59, January 1993.
[3]L. P. Russo and R. E. Young, “Moving-horizon state estimation applied to an industrial polymerization process,” in Proc. 1999 American Control Conference, vol. 2, San Diego, California, USA, 1999, pp. 1129–1133.
[4]E. L. Haseltine and J. B. Rawlings, “Critical evaluation of extended Kalman filtering and moving-horizon estimation,” 2005 American Chemical Society, vol. 44, no. 8, pp. 2451–2460, 2005.
[5]A. Alessandri, M. Baglietto, and G. Battistelli, “Moving-horizon state estimation for nonlinear discrete-time systems: New stability results and approximation schemes,” Automatica, vol. 44, no. 7, pp. 1753–1765, July 2008.
[6]T. A. Johansen, D. Sui, and R. Nybø, ”Regularized nonlinear moving-horizon observer with robustness to delayed and lost data,” IEEE Transactions on Control Systems Technology, vol. 21, no.6, pp. 2114–2128, Nov. 2013.
[7]M. Canale, L. Fagiano, and C. Novara, “A DVS-MHE Approach to Vehicle Side-Slip Angle Estimation,” IEEE Transactions on Control Systems Technology, vol. 22, no. 5, Sept. 2014, pp. 2048–2055.
[8]B. D. O. Anderson and J. B. Moore, Optimal Filtering, Dover Publications, Inc., 2005.
[9]M. Mohandes, S. Rehman and T. Halawani, “A neural networks approach for wind speed prediction,” Renewable Energy, vol. 13, no. 3, pp. 345–354, March 1998.
[10]G. Li and J. Shi, “On comparing three artificial neural networks for wind speed forecasting,” Applied Energy, vol. 87, no. 7, pp.2313–2320, July 2010.
[11]Y. Wang, D. L. Wu, C. X. Guo, Q. H. Wu, W. Z. Qian and J. Yang, “Short-term wind speed prediction using support vector regression,” in Proc. Power and Energy Society General Meeting, July 2010.
[12]M. A. Mohandes, T. O. Halawani, S. Rehman and A. A. Hussain, “Support vector machines for wind speed prediction,” Renewable Energy, vol. 29, no. 6, pp. 939–947, May 2004.
[13]E. G. Ortiz-García, S. Salcedo-Sanz, Á. M. Pérez-Bellido, J. Gascón-Moreno, J. A. Portilla-Figueras and L. Prieto, “Short-term wind speed prediction in wind farms based on banks of support vector machines,” Wind Energy, vol. 14, no. 2, pp. 193–207, March 2011.
[14]S. Li, “Wind power prediction using recurrent multilayer perceptron neural networks,” in Proc. IEEE Power Engineering Society General Meeting, 2003, pp. 2325–2330.
[15]A. Răzvan-D, F. P.-Vlădicescu and G. Kalmuțchi, “A recurrent neural network approach for short term wind power prediction,” Journal of Sustainable Energy, vol. 2, no. 4, Dec. 2011.
[16]S. Balluff , J. Bendfeld and S. Krauter, “Short term wind and energy prediction for offshore wind farms using neural networks,” in Proc. 2015 International Conference on Renewable Energy Research and Applications (ICRERA), Feb. 2016.
[17]E. A. Bossanyi, “Short-term wind prediction using Kalman filters,” Wind Engineering, vol. 9, no. 1, pp. 1–8, 1985.
[18]T. Ishikawa and T. Namerikawa, “Short-term wind power prediction for wind turbine via kalman filter based on JIT modeling,” in Proceedings of SICE Annual Conference (SICE), September 2013.
[19]C. D. Zuluaga, M. A. Álvarez and E. Giraldo, “Short-term wind speed prediction based on robust Kalman filtering: An experimental comparison,” Applied Energy, vol. 156, pp. 321–330, 2015.
[20]K. Chen and J. Yu, “Short-term wind speed prediction using an unscented Kalman filter based state-space support vector regression approach,” Applied Energy, vol. 113, pp. 690–705, Jan. 2014.
[21]M. C. Alexiadis, P. S. Dokopoulos and H. S. Sahsamanoglou, “Wind speed and power forecasting based on spatial correlation models,” IEEE Transactions on Energy Conversion, vol. 14, pp. 836–842, Sep. 1999.
[22]D. J. Nott, W. T. M. Dunsmuir, R. Kohn and F. Woodcock, “Statistical correction of a deterministic numerical weather prediction model,” Journal of The American Statistical Association, vol. 96, no. 455, pp.794–804, Sep. 2001.
[23]L. Landerg, G. Giebel, H. A. Nielsen, T. Nielsen and H. Madsen, “Short-term prediction - an overview,” Wind Energy, vol. 6, no. 3, pp. 273–280, June 2003.
[24]P. Lynch, “The origins of computer weather prediction and climate modeling,” Journal of Computational Physics, vol. 227, no. 7, pp. 3431–3444, March 2008.
[25]M. Lange and U. Focken, “New developments in wind energy forecasting,” in Proc. 2008 IEEE Power and Energy Society General Meeting, Pittsburgh, PA, July 2008.
[26]B. G. Brown, R. W. Katz and A. H. Murphy, “Time series models to simulate and forecast wind speed and wind power,” Journal of Climate and Applied Meteorology, vol. 23, pp. 1184–1195, May 1984.
[27]Y. Kassa, J. H. Zhang, D. H. Zheng and D. Wei, “Short term wind power prediction using ANFIS,” in Proc. IEEE International Conference on Power Power and Renewable Energy, March 2017.
[28]H. M. I. Pousinho, V. M. F. Mendes and J. P. S. Catalão, “A hybrid PSO–ANFIS approach for short-term wind power prediction in Portugal,” Energy Conversion and Management, vol. 52, no. 1, pp. 397–402, Jan. 2011.
[29]D. Zheng, A. T. Eseye, J. Zhang and H. Li, “Short-term wind power forecasting using a double-stage hierarchical ANFIS approach for energy management in microgrids,” Protection and Control of Modern Power Systems, vol. 2, no. 13, Dec. 2017.
[30]E. A. WAN, R. Van Der Merwe, “The unscented Kalman filter for nonlinear estimation,” in Proc. The IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium, 2000, pp. 153–158.
[31]M.-C. Wu, J.-S. Hong, L.-F. Hsiao, L.-H. Hsu and C.-J. Wang, “Effective use of ensemble numerical weather predictions in Taiwan by means of a SOM-based cluster analysis technique,” Water, vol. 9, no. 11, pp. 836, October 2017.
[32]V. N. Vapnik, The Nature of Statistical Learning Theory, Springer-Verlag, New York, 1995.
[33]V. N. Vapnik, Statistical Learning Theory, John Wiley & Sons, New York, 1998.
[34]B. Schölkoph and A. J. Smola, Learning with kernels, MIT Press, Cambridge, MA, 2002.
[35]C. -J. Lin, “Formulations of support vector machines: a note from an optimization point of view,” Neural Computation, 13(2):307–317, 2001.
[36]J. Shawe-Taylor and N. Cristianini, Kernel methods for pattern analysis, MIT Press, Cambridge, MA, 2004.
[37]J. B. Rawlings, D. Q. Mayne and M. M. Diehl, “Model Predictive Control: Theory, Computation, and Design,” 2nd Ed., Nob Hill Publishing, LLC, Oct. 2017.
[38]K. R. Muske, J. B. Rawlings and J. H. Lee, “Receding Horizon Recursive State Estimation,” in Proc. American Control Conference, San Francisco, CA, June 1993, pp. 900–904.
[39]C. V. Rao, J. B. Rawlings and J. H. Lee, “Constrained linear state estimation - a moving horizon approach,” Automatica, vol. 37, no. 10, pp. 1619–1628, Oct. 2001.
[40]C.V. Rao, J.B. Rawlings and D.Q. Mayne, “Constrained state estimation for nonlinear discrete-time systems: stability and moving horizon approximations,” IEEE Transactions on Automatic Control, vol. 48, Feb. 2003, pp. 248–258.
[41]S. S. Keerthi and E. G. Gilbert, “Optimal infinite-horizon feedback laws for a general class of constrained discrete-time systems: Stability and moving-horizon approximations,” Journal of Optimization Theory and Applications, vol. 57, no. 2, pp. 265–293, May 1988.
[42]S. Ungarala, “Computing arrival cost parameters in moving horizon estimation using sampling based filters,” Journal of Process Control, vol. 19, no. 9, pp. 1576–1588, Oct. 2009.