本研究的目的是建立有限元素分析模型，模擬雷射玻璃彎曲的過程。利用ANSYS APDL軟體建立有限元素模型，由於玻璃試片很薄，因此使用殼元素。在模擬中，使用了隨溫度變化的材料參數，並利用廣義的Maxwell模型來模擬玻璃的黏彈性應力鬆弛現象。 在實驗中，使用波長為1070 nm的連續波光纖雷射照射0.5 mm厚的鈉鈣玻璃。為了驗證有限元素模型的有效性，進行單點雷射與移動式雷射照射實驗，並使用紅外線測溫儀量測玻璃試片的表面溫度。實驗中，總共七個位置的量測溫度隨時間變化均與模擬計算結果相符。 在雷射彎曲中，模擬了三種掃描策略及每種掃描策略中兩種不同的雷射功率，並與實驗結果進行比較。根據模擬結果，玻璃的機械強度估計約為 224-230 MPa，雖然它高於實際玻璃的機械強度，但模擬與實驗的趨勢是一致的。模擬結果顯示，在雷射掃描過程中，產生較高絕對最大主應力的掃描策略下會發生玻璃斷裂，實驗也證明了這一點。模擬中計算的玻璃彎曲角度小於實驗量測的彎曲角度，這是因為模擬中，所使用的玻璃材料機械性質參數，在高溫區域不甚充足。 對於在試片寬度方向上進行單一雷射掃描的掃描策略，隨著雷射光斑靠近模型邊緣，邊緣會產生極高的拉伸應力，導致試片破裂。先預熱試片邊緣後再掃描整個試片寬度的掃描策略，會讓絕對最大主應力比前一個掃描策略低很多。第三種策略是先預熱試片邊緣後再在試片寬度內掃描，這種掃描策略會讓邊緣點的應力在整個彎曲的過程中都保持較低。依此，就能成功利用雷射完成玻璃彎曲。;The purpose of this study is to develop a computer-aided-engineering (CAE) technique using finite element method (FEM) to simulate the process of laser bending of glass. The commercial FEM code, ANSYS APDL, is applied to develop the FEM model. In the simulation, shell element is used since the glass specimen is very thin. Temperature-dependent material properties are considered. The viscoelastic stress relaxation behavior is considered in this study and is described by a generalized Maxwell model. In validating experiments, a 500-W fiber laser of continuous wave with a wavelength of 1070 nm is used to irradiate the soda-lime glass of 0.5 mm in thickness. Infrared thermometer is used to measure the surface temperature of the glass specimen to validate the effectiveness of the FEM model. Two kinds of laser irradiation experiments are conducted, including fixed-point laser irradiation and moving laser irradiation. Totally seven locations are selected to measure the temperature. All temperature variations at selected locations calculated in the simulation agree well with the experiments. For laser bending, three kinds of scanning strategies and two different laser powers in each scanning strategy are simulated and compared with the experimental results. The mechanical strength of the given glass is estimated to be around 224-230 MPa in the simulation. Although it is higher than the actual mechanical strength of the given glass, trends in both simulation and experiment are consistent. Early breakage occurs in the cases with a higher absolute maximum principal stress, also evidenced by the experiments. The calculated bending angle is smaller than that measured in experiment, which may be attribute to insufficient high-temperature mechanical properties in simulation. For the scanning strategy with a single laser scanning track across the specimen width, high tensile stress is generated at the edge of the model as the laser spot moves close to the edge. This causes the failure of laser bending. For scanning strategy with preheating the edges first followed by scanning across the specimen width, the absolute maximum principal stress is much lower than the previous strategy for a similar laser power. For the third laser scanning strategy which preheats the edges and then scans within the specimen width, the stress of the edge point remains low throughout the entire laser bending process. In this way, the glass specimen is successfully bent by laser processing.
