本研究擬以參數設計最佳化概念來進行內模式控制器(Internal Model Control, IMC)之最佳設計參數分析，內模式控制器(IMC)是一種以數學模型進行推衍控制器設計的控制策略，內模式控制器(IMC)之設計優勢其結構簡單、設計直觀，其設計參數容易調整且易於實現等特性。本研究所使用閉迴路控制系統，在內模式控制架構中，具有轉移函數、受控體、受控體近似模型，它經由系統方塊之簡化可轉換成等效閉迴路控制系統，本研究之受控體(本船模式)採用Nomoto 一階模式，此一階模式係利用原Nomoto 二階模式簡化而成，利用IMC 不穩定系統設計流程經過公式推導，產生控制器通式，控制器參數中，以船舶迴旋性測試所得之相對應的舵角增益k、時間常數T 數值代表本船特性，IMC 設計參數β值為本次研究分析的主要設計參數，其中β值大小代表控制器的反應時間快慢，經由研究試驗可找出控制器設計時之最適化β值。經由研究分析結果顯示，當設計參數β值過小時，整體系統反應容易發生震盪現象，當β值較小，船舶達到指定航向角之反應時間相對較短，但其超越量則較大達到穩態反應時間較短，反之當設計之β值較大時，船舶達到指定航向角時間較長，亦即船舶達到穩態之反應時間相對較長。

This research intends to adopt the concept of parameter design optimization to find out the design parameter β value for the internal model control (IMC) method. The internal model control (IMC) method adopted in this paper belongs to the model-based approach, which is characterized by explicit dependence between the plant model parameters and the controller parameters. In this study the simplicity of the 1st order Nomoto model renders it an ideal basis for an autopilot design. In the 1st order Nomoto model, parameter T is the effective yaw mode time constant and k is the rudder gain. The IMC structure, the controller gains depend explicitly on the plant parameters k、T and the design parameter β. Once the plant model is chosen and the design parameter β is selected, the controller follows immediately. The IMC structure can be transformed into the classical feedback structure. In the internal model control (IMC) structure the β is a design parameter that determines the speed of response of the closed-loop system. Moreover, the IMC controller is intuitively straightforward and easily implemented using existing software. Most of the control engineers were engaged in design a prefect control system which with fast response and without any overshoots. In this study intends to use the parameter design optimization technique to find the best design parameter β to achieved both quick response time and without overshoots at same time. The research and analysis results show that when the design parameter β value is too small then the overall system response is prone to oscillating. When the β value is small, the response time of the ship to reach the specified heading angle is relatively fast, but its sacrifice the overshoot. Conversely, when the β value is large, the time required for the vessel to reach the specified heading angle is longer to reach steady state, that is, the response time for the vessel to reach steady state is relatively long.