本文發展了一套結合類神經網路自調式PD控制器之船舶運動數學模式來模擬控制拖船之角度，藉調整拖船之拖曳角度的改變而產生不同的分力作用於被拖船上，以達到被拖船航線與航向之控制。本文中考慮在規則波中以不同的波向及不同的波長下做探討，並且加上風力及漂移力的影響下，以類神經網路自動控制來模擬被拖船之航向軌跡及運動狀態。由模擬計算結果發現本文發展之類神經網路自調式PD控制器，一般的確可達到較佳且快速之控制並節省了許多傳統PD控制尋找最佳化增益參數所消耗之時間。因此本文建立之數學模式，對於海上或限制水域之拖船系統控制評估可提供有用之參考價值。

The present paper develops a ship motion mathematical model including the neural network self-tuning PD control to simulate the towing angle of the tugboats. Through the automatic control on the towing angle, the tracking trajectory and course keeping of the towed ship can be handled well. The towing system is considered in regular wave with different headings and frequencies. The effects due to the wind force and wave drift are also included. From the present simulation results, we find that the neural network self-tuning PD controller developed here indeed can achieve a better and quick control on the course keeping of the towed ship and save the consuming time for searching the optimal control gain which the conventional PD controller can not do. Therefore the mathematical model including the neural network self-tuning PD control can offer a valuable reference to the towing system control at sea or confined water area.

1. Crossland, P. and Johnson, M.C., “A time domain simulation of deck wetness in head seas,” Proceeding of RINA International Conference on Ship Motions and Manoeuverability, London, UK, 1998.
2. Fang, M.C., Lee, M.L., and Lee, C.K., “The simulation of water shipping for a ship advancing in large longitudinal waves,” Journal of Ship Research, Vol. 37, pp. 26-137, 1993.
3. Grino, R., Cembrano, G. and Torras, C., ” Nonlinear System Identification Using Additive Dynamic Neural Networks—Two On-Line Approaches,” IEEE Transactions on Circuits andSystems I: Fundamental Theory and Applications, Vol. 47,No. 2, pp. 150-165, February 2000.
4. Hamamoto, M. and Kim, Y.S., “A new coordinate system and the equations describing maneuvering motion of a ship in waves (in Japanese),” Journal of the Society of Naval Architects of Japan, Vol. 173, pp. 209-220, 1993.
5. Hamamoto, M., Matsuda, A., and Ise, Y., “Ship motion and the dangerous zone of a ship in severe following seas (in Japanese),” Journal of the Society of Naval Architects of Japan, Vol. 175, pp. 69-78, 1994
6. Hemerly, E. M. and Nascimento, Jr, C. L. ”An NN-basedapproach for tuning servocontrollers,” Neural Networks,Vol. 12, No. 3, pp. 513-518, April 1999.
7. Hirano, M., “On the calculation method of ship maneuvering motion at initial design phase (in Japanese),” Journal of the Society of Naval Architects of Japan, Vol. 147, pp. 144-153, 1980.
8. Inoue, S., Hirano, M., Kijima, K., and Takashina, J., “A practical calculation method of ship maneuvering motion,” International Shipbuilding Progress, Vol. 28, pp. 207-222, 1981.
9. Isherwood, R.M., “Wind Resistance of Merchant Ships,” Transactions of RINA, Vol. 115, pp. 327-338, 1973.
10. Kishimoto, T. and Kijima, K., “On the Towing System for Disabled Ships(2nd Report),” Journal of the Society of Naval Architects of Japan, Vol.191, pp. 77-85, 2002.
11. Rumelhart, D. E. and McClelland, J. L.,”Parallel Distributed Processing: Explorations in the Microstructure of Cognition.” Vol. 1. Cambridge, MA:MIT Press, 1986.
12. Yoshimura, Y. and Nomoto, K., “Modeling of manoeuvring behavior of ships with a propeller idling, boosting and reversing (in Japanese),” Journal of the Society of Naval Architects of Japan, Vol. 144, pp. 57-69, 1978.
13. 關西造船協會，造船設計便覽(日文)，海文堂出版株式會社，東京，1983。
14. 高品純志，港灣域における船の操縱運動計算法に關する研究(日文)，博士論文，東京大學，日本，1992。
15. 張斐章，張麗秋，類神經網路，東華書局，民國九十五年。
16. 葉怡成，類神經網路模式應用與實作，儒林圖書有限公司，民國九十五年。
17. 林緯作，環境外力下船舶拖帶模擬計算系統之研發與應用，國立台灣大學工程科學及海洋工程學研究所，民國九十四年。

18. 徐玉樹，順浪中船舶之穩度安全性，國立成功大學造船暨船舶機械工程研究所碩士論文，民國八十六年。
19. 羅志宏，船隻在順波中航向穩定性與運動之分析，國立成功大學造船暨船舶機械工程研究所碩士論文，民國九十年。
20. 羅志宏，船舶在波浪中之非線性運動操控模式之探討，國立成功大學造船暨船舶機械工程研究所博士論文，民國九十五年。
21. 卓永烱，應用類神經網路於船舶減搖控制器之研究，國立成功大學系統及船舶機電工程研究所碩士論文，民國九十七年。