肋為CAD模型中常見的特徵，在零件設計中通常用來作為補強的功用。在CAE分析中，需要由CAD模型轉換為網格，由CAD模型中擷取肋的特徵有助於網格化方法的發展，以便建立合適的網格。本研究的目的為發展CAE網格化應用之肋特徵辨識技術。在本方法中，需要建立邊與面的拓樸資料，我們使用AAG的方法建立邊與面的關聯性資訊。此外，在CAD模型中常有混接曲面，使得邊與面的關係更複雜，因此，混接面之拓樸資料也需要建立並加以應用。本研究使用的搜尋方法為rule-based的方法，建立一組規則用以搜尋屬於肋的面。本方法的完整流程將在論文中完整說明，流程中每一步驟也將詳細說明。為驗證本研究方法的可行性，使用超過30個CAD模型作為測試案例，所有結果並整理成表格與討論。這些範例中絕大多數的肋均可正確地辨識，少數錯誤判斷的情況與其原因也將於論文中說明。;Rib is a typical feature in CAD models. It is generally used to provide extra strength for the product design. In CAE analysis, it is required to convert CAD models into meshes. The extraction of ribs from a CAD model can help the development of the meshing algorithm for generating appropriate meshes. The purpose of this study is to develop a feature recognition algorithm for ribs for the application in CAE meshing process. In the proposed algorithm, the topological data among edges and faces must be established. We employ the AAG (attributed adjacency graph) method to record topological information among edges and faces. Moreover, fillets often appear on the CAD model and complicate the adjacent relationship among edges and faces. Therefore, a topological data related to fillets must also be established and employed. The search method used in the proposed algorithm is a rule-based method, in which a set of rules are provided to determine the faces which belong to a rib. The overall flowchart of the proposed method is described in this thesis, and a detailed description for each of the steps is explained. To verify the feasibility of the proposed method, more than 30 CAD models are tested and results are listed in tables and discussed. Most of the ribs on these cases can be detected accurately. The situations which may result in erroneous recognition of ribs for limited cases are discussed.
