在凝固熱傳過程中，固態與液態之間的相變化是相當重要的物理現象，能夠節省計算時間的可變時間步伐計算法亦有不少種。本文針對兩個較為普遍的可變時間計算步伐來作探討，一個為GLS法，另一為修正 GLS（MGLS）法，是針對GLS法再進行修正所提出的計算法。本研究以這兩種可變時間步伐計算法搭配數種處理潛熱的數值方法，來求解不同凝固熱傳過程的溫度場分布，並對總時間步伐數進行分析比較。本文將介紹GLS法與MGLS法的計算理論，及處理潛熱效應的數值方法，包含等效比熱與熱焓法。所計算的凝固問題有一維史蒂芬問題、紐曼問題、二維瑞特延問題和二維鑄模之凝固問題，上述問題除了二維鑄模之凝固問題之外，其數值解皆與解析解進行準確性和誤差比較，再與固定步伐進行總步伐數的分析，而求得最合適的步伐數。分析的結果發現，在求解史蒂芬問題與紐曼問題時，熱焓法在處理潛熱效應時能提供最好的準確性，而MGLS法確實能比GLS法來得更有效率，提供較少的總步伐數。從這些分析結果，針對凝固熱傳問題，可發現本文的方法可以有效的提供精簡且合適的時間步伐，以在不犧牲準確度的前提下達到節省計算時間的目的。

Phase change is an important physical phenomenon in a solidification process. There are many adaptive time step methods proposed for saving the computing time. In the study, to analyze nonlinear phase-change or solidification processes with two numerical schemes of handing the latent-heat effect, two adaptive time step methods are chosen, one is the Gresho-Lee-Sani (GLS) predictor-corrector strategy and the other is the modified GLS (MGLS) method. The temperature fields of different solidification problems are solved and their total time step numbers are compared and analyzed. In the thesis, the computing algorithms of GLS and MGLS methods are presented and so are the latent-heat treating methods, including the effective specific heat and enthalpy schemes. The computational accuracy and efficiency of the two adaptive time step methods are demonstrated via the simulation of one-dimensional Stefan and Neumann phase-change problems, the two-dimensional Rathjen phase-change problem and the solidification problem of a casting process. In those problems except the two-dimensional one of casting process, the accuracy is compared by analyzing the differences between the numerical and exact solutions for the adaptive and uniform time step methods and so is the total step number. From the analyzing results, it can be found that the enthalpy scheme could give the more accurate solutions than the effective specific heat method for the Stefan and Neumann problems. The adaptive method could solve the problems more efficiently than the uniform one and the MGLS scheme could do better than the GLS one.

[1] Gaskell PH, Jimack PK, Sellier M and Thompson HM. Efficient and accurate time adaptive multigrid simulations of droplet spreading. International Journal for Numerical Methods in Fluids 2004; 45: 1161-1186.
[2] Kavetski D, Binning P and Sloan SW. Adaptive backward euler time stepping with truncation error control for numerical modelling of unsaturated fluid flow. International Journal for Numerical Methods in Engineering 2002; 53: 1301-1322.
[3] Kavetski D, Binning P and Sloan SW. Adaptive time stepping and error control in a mass conservative numerical solution of mixed form of Richards equation. Advances in Water Resources 2001; 24: 595-605.
[4] Crouzet N and Turinsky PJ. A second-derivative-based adaptive time-step method for spatial kinetics calculation. Nuclear Science Engineering 1996; 123: 206-214.
[5] Chen K, Baines MJ and Sweby PK. On an adaptive time stepping strategy for solving nonlinear diffusion equations. Journal of Computational Physics 1993; 150: 324-332.
[6] Dantzig JA. Modeling liquid-solid phase changes with melt convection. International Journal for Numerical Methods in Engineering 1989; 28: 1769-1785.
[7] Douglas J and Gallie TM. On the numerical integration of a parabolic differential equation subject to a moving boundary condition. Duke Mathematical Journal 1955; 22: 557-571.
[8] Goodling JS and Khader MS. Inward solidification with radiation-convection boundary condition. Journal of Heat Transfer-Transactions of the ASME 1974; 96: 114-115.
[9] Gupta RS and Kumar D. A modified variable time step method for one-dimensional Stefan problem. Computer Methods in Applied Mechanics and Engineering 1980; 23: 101-109.
[10] Gupta RS and Kumar D. Variable time step methods for one-dimensional Stefan problem with mixed boundary condition. International Journal of Heat and Mass Transfer 1981; 24: 251-259.
[11] Ouyang T and Tamma KK. On adaptive time stepping approaches for thermal solidification processes. International Journal of Numerical Methods for Heat and Fluid Flow 1996; 6: 37-50.
[12] Gresho PM., Lee RL and Sani RL. Recent Advance in Numerical Methods in Fluids. Pineridge Press: Swansea, 1980.
[13] FIDAP Theory Manual. Fluent Inc.: New Hampshire, 1998.
[14] Hu H and Argyropoulos SA. Mathematical modeling of solidification and melting: a review. Modelling and Simulation in Materials Science and Engineering 1996; 4: 371-396.
[15] Voller VR and Swaminathan CR. Fixed grid techniques for phase change problems: a review. International Journal for Numerical Methods in Engineering 1990; 30: 875-898.
[16] Tamma KK and Namburu RR. Recent advances, trends and new perspectives via enthalpy-based finite element formulations for applications to solidification problems. International Journal for Numerical Methods in Engineering 1990; 30: 803-820.
[17] Namburu RR and Tamma KK. Effective modeling/analysis of isothermal phase change problems with emphasis on representative enthalpy architectures and finite elements. International Journal of Heat and Mass Transfer 1993; 36: 4493-4497.
[18] Carslaw HS and Jaeger JC. Conduction of Heat in Solids. Oxford University Press: Oxford, 1959.
[19] Rathjen KA and Jiji LM. Heat conduction with melting or freezing in a corner. Journal of Heat Transfer-Transactions of the ASME 1971; 93: 101-109.
[20] Hsiun-Chang Peng and Long-Sun Chao. A modified truncation error-based adaptive time-step control strategy for fully and semi-implicit simulation of isothermal solidification processes. International Journal of Thermal Sciences 2009 ; 48: 1517–1522.
[21] Hsiun-Chang Peng and Long-Sun Chao. Adaptive time stepping strategy for solidification processes based on modified local time truncation error. International Journal for numerical methods in engineering 2009; 79:1245–12.