新穎的非等溫熱成型和冷模淬火（HFQ®）工藝可以經濟高效地生産複雜形狀的高強度鋁合金面闆元件。但是，由于對新工藝的設計不熟悉，妨礙了其在工業環境中的廣泛采用。最近的研究工作集中于開發用于有限元模拟的進階材料模型，該模型用于評估HFQ®工藝新元件設計的可行性。但是，有限元模拟在設計過程的後期進行，需要形成過程的專業知識，不适合早期的設計探索。為了解決這些局限性，本研究提出了一種基于卷積神經網絡（CNN）的替代産品的新應用，該替代産品可以快速評估使用HFQ®工藝形成的零件的制造可行性。包含元件幾何形狀，毛坯形狀和加工參數的變化以及相應的實體場的多樣化資料集被生成并用于訓練模型。結果表明，與HFQ®仿真相比，該模型實時獲得了幾乎無法區分的全場預測。該技術為在HFQ®條件下形成的複雜形狀的零件的設計過程開始時提供了寶貴的工具，以幫助零件設計和決策。
原文題目：Rapid feasibility assessment of components formed through hot stamping: A deep learning approach
原文：The novel non-isothermal Hot Forming and cold die Quenching (HFQ) process can enable the cost-effective production of complex shaped, high strength aluminium alloy panel components. However, the unfamiliarity of designing for the new process prevents its widescale adoption in industrial settings. Recent research efforts focus on the development of advanced material models for finite element simulations, used to assess the feasibility of new component designs for the HFQ process. However, FE simulations take place late in design processes, require forming process expertise and are unsuitable for early-stage design explorations. To address these limitations, this study presents a novel application of a Convolutional Neural Network (CNN) based surrogate as a means of rapid manufacturing feasibility assessment for components to be formed using the HFQ process. A diverse dataset containing variations in component geometry, blank shapes, and processing parameters, together with corresponding physical fields is generated and used to train the model. The results show that near indistinguishable full field predictions are obtained in real time from the model when compared with HFQ simulations. This technique provides an invaluable tool to aid component design and decision making at the onset of a design process for complex-shaped components formed under HFQ conditions.
熱沖壓成形零件的快速可行性評估：一種深度學習方法.pdf