This work investigates the performance of the on-demand machine learning (ODML) algorithm introduced in Leal et al. (2020) when applied to different reactive transport problems in heterogeneous porous media. ODML was devised to accelerate the computationally expensive geochemical reaction calculations in reactive transport simulations. We demonstrate that the ODML algorithm speeds up these calculations by one to three orders of magnitude. Such acceleration, in turn, significantly accelerates the entire reactive transport simulation. The numerical experiments are performed by implementing the coupling of two open-source software packages: Reaktoro (Leal, 2015) and Firedrake (Rathgeber et al., 2016).

中文翻译：

---

使用 Reaktoro 和 Firedrake 在异质多孔介质中加速反应输运模拟

这项工作调查了 Leal 等人引入的按需机器学习 (ODML) 算法的性能。(2020) 当应用于非均质多孔介质中的不同反应传输问题时。ODML 旨在加速反应输运模拟中计算成本高昂的地球化学反应计算。我们证明了 ODML 算法将这些计算速度提高了一到三个数量级。这种加速反过来又显着加速了整个反应输运模拟。数值实验是通过实现两个开源软件包的耦合来执行的：Reaktoro（Leal，2015）和 Firedrake（Rathgeber 等，2016）。