本研究中,使用含蒸發系統之1kW甲醇蒸氣重組器,且對商業MDC-3觸媒進行甲醇蒸氣重組反應之實驗研究,採用均勻設計(Uniform Design)方法以及接續式擬均勻設計(Sequential Pseudo-Uniform Design)方法佈置實驗點的輸入值(水對甲醇莫耳比[H2O]/[CH3OH] γ、甲醇進料量 和反應溫度 T)。以總質量平衡為根據,經由量測入口端進料和出口端冷凝液的密度,可以獲得反應中各成分平衡所需的數值。 以經驗法則和高氫氣產率以及低觸媒老化所得到的目標方程式為根據來逐步縮小實驗的操作範圍,如此便能得到可以接受的1kW甲醇蒸氣重組器之最佳操作條件。進行安排的39組實驗中,能獲得甲醇蒸氣重組器的最佳操作條件為γ = 1, T = 557 (K), = 0.383 (mol/min)。以上所敘述的方法,可以不需經由模式便可獲得一個未知程序的最佳操作條件以及驗證在本反應器實驗系統是可行的。
In this study, a 1kW methanol steam reformer including an evaporating system was set up to carry out the experimental study of the methanol steam reforming reaction adopting the commercial MDC-3 catalyst. The uniform design (UD) method and the sequential pseudo-uniform method (SPUD) method were adopted to locate the inputs of the experiments (the mole ratio of [H2O]/[CH3OH] γ, the methanol molar feed rate , and the reacting temperature T). On the basis of the overall mass balance, the component balance for each species involved in the reactions can be obtained via the density measurement of the input and output liquid fluid flows. The most acceptable optimizing operating condition for running the 1kW constructed steam reformer was achieved by the method of contracting the promising operating region. Based on the proposed heuristic algorithm and the given objective function for achieving high yield of hydrogen and low catalyst deactivation, the most acceptable optimizing operating condition γ = 1, T = 557 (K), =0.383 (mol/min) was obtained through a total 39 experiments. By this way, a model-free approach to optimizing an unknown process was developed and proven to be applicable in this experimental reactor system.