The deep learning boom motivates researchers and practitioners of computational fluid dynamics eager to integrate the two areas.The PINN (physics-informed neural network) method is one such attempt. While most reports in the literature show positive outcomes of applying the PINN method, our experiments with it stifled such optimism. This work presents our not-so-successful story of using PINN to solve two fundamental flow problems: 2D Taylor-Green vortex at $Re = 100$ and 2D cylinder flow at $Re = 200$. The PINN method solved the 2D Taylor-Green vortex problem with acceptable results, and we used this flow as an accuracy and performance benchmark. About 32 hours of training were required for the PINN method's accuracy to match the accuracy of a $16 \times 16$ finite-difference simulation, which took less than 20 seconds. The 2D cylinder flow, on the other hand, did not even result in a physical solution. The PINN method behaved like a steady-flow solver and did not capture the vortex shedding phenomenon. By sharing our experience, we would like to emphasize that the PINN method is still a work-in-progress. More work is needed to make PINN feasible for real-world problems.
翻译:深层的学习繁荣激发了研究人员和计算流动态从业人员渴望将这两个区域融合起来。 PINN(物理-知情神经网络)方法就是这样一种尝试。虽然文献中的大多数报告显示应用 PINN 方法的积极成果,但我们的实验窒息了这种乐观。 这项工作展示了我们使用 PINN 解决两个基本流量问题的不成功的故事: 2D Taylor-Green 涡轮以 $re = 100美元和 2D 气瓶流以 $ Re = 200 美元。 PINN 方法以可接受的结果解决了 2D Taylor- Green 旋涡问题,我们用这个方法作为精确和性能基准。 PIN 方法的准确性需要大约32小时的培训,以达到16 y time 16 $ 的有限波动模拟的准确性,这花了不到20 秒。 2D 气瓶流甚至没有以实物方式解决。 PIN 方法像一个稳定的流流流解者一样,没有捕捉到 Pvoltex-N 方法, 通过分享我们真正的工作需要更多的进展。