本研究的目的為以數值方法計算漁船凍藏艙艙壁厚度的最佳化，期能提昇造船廠對凍藏艙的設計能力， 以達到降低成本及提高造船品質的成效。研究架構首先以SolidWorks建立凍藏艙模型，輸入AnsysWorkbench 後產生網格系統，再以AnsysWorkbench穩態熱傳導分析軟體計算艙壁的熱傳導量，結合多目標粒子群最佳 化法(Multi-Objective Particle Swarm Optimization，MOPSO)後可求得最佳化的隔熱層配置；前述的最佳化 設計系統以高雄某造船廠設計圖為測試實例，結果顯示在保持原設計的隔熱效果條件下，可節省約3.6%的 隔熱材料使用量，亦或使用原廠相同的隔熱材使用量，則最佳的配置可減少4.7%的熱流入量，計算實例顯 示了最佳化設計系統的有效性，隔熱材料的減量使用亦能呼應國內外環保的趨勢。

The research aims to developing a numerical method to optimize the frozen tank of fishing vessels. The application of the method can enhance the design ability of shipyards so as to lower down the cost and lift up the construction quality. The numerical method involves three parts. SolidWorks is used to build up the model of the frozen tank. Ansys is used to create the grid system and carry out the calculation of heat transfer problem. Finally, MOPSO is used to search for the Pareto front. The numerical method is examined by taking an original design of a shipyard in southern Taiwan. The results indicate that the amount of insulation material decreases 3.6% when keeping the same heat transfer condition. The amount of heat transfer decreases 4.7% when using the same quantity of insulation material. Numerical results demonstrate the availability of the proposed numerical method. The reduction of insulation material is also beneficial to the environmental protection.