The Kampmann–Wagner Numerical (KWN) model of precipitation is a powerful tool to simulate the precipitation of the second phase considering the nucleation, growth, and coarsening. Some quantities such as interfacial energy and nucleation site number density are required to accomplish the simulation. Practically, those quantities are hard to measure in the experiment directly, and the derivation of those quantities through modeling can also be costly. In this work, we hereby adopt the minimization algorithm implemented in the open-source Scipy Python package to derive that important information in terms of very limited experimental data. The convergence and robustness of different algorithms are discussed. Among those algorithms, the Nelder–Mead and Powell algorithms are successfully applied to optimize multiple parameters during KWN modeling. This work will shed light on the design of experiments/processes and facilitate integrated computational materials engineering (ICME).

Kampmann-Wagner Numerical （KWN） 降水模型是模拟第二阶段降水的强大工具，考虑了成核、生长和粗化。完成仿真需要一些物理量，例如界面能和成核位点数密度。实际上，这些量很难在实验中直接测量，而且通过建模推导这些量也可能很昂贵。在这项工作中，我们特此采用开源 Scipy Python 包中实现的最小化算法，以非常有限的实验数据得出重要信息。讨论了不同算法的收敛性和鲁棒性。在这些算法中，Nelder-Mead 和 Powell 算法成功地应用于 KWN 建模过程中的多个参数优化。这项工作将阐明实验/过程的设计，并促进集成计算材料工程 （ICME）。