Physics-guided machine learning from simulated data with different physical parameters
Physics-guided machine learning from simulated data with different physical parameters
复制标题
DOI：
10.1007/s10115-023-01864-z
复制
发表时间：
2023-03-31
影响因子：
2.7
通讯作者：
Jia, Xiaowei
中科院分区：
文献类型：
--
作者：
Chen, Shengyu;Kalanat, Nasrin;Jia, Xiaowei
基于物理的模型被广泛应用于研究各种科学与工程问题中的动态系统。然而，由于对潜在过程建模时知识不完备或过程过于复杂，这些模型必然只是对现实的近似。因此，由于用于表征真实物理过程的参数化不准确或近似不当，它们常常产生有偏差的模拟结果。在本文中，我们旨在构建一个全新的物理引导机器学习框架来监测动态系统。其思路是利用先进的机器学习模型提取复杂的时空数据模式，同时融入基于物理的模型所生成模拟数据中体现的一般科学知识。为应对因参数化不完善而导致的模拟数据偏差，我们提议从基于物理的模型在不同物理参数下生成的多组模拟数据中联合提取一般物理关系。具体而言，我们开发了一种时空网络架构，该架构利用其门控变量捕捉物理参数的变化。我们采用预训练策略对该模型进行初始化，这有助于发现不同组模拟数据共有的常见物理模式。然后，结合有限的观测数据和充足的模拟数据对其进行微调。通过利用机器学习与领域知识的互补优势，我们的方法已被证明能够做出准确预测、使用更少的训练样本，并能推广应用于样本外的场景。我们进一步表明，该方法在两个领域应用中——预测溪流温度和预测湖泊温度——能够深入洞察物理参数随空间和时间的变化情况。
Physics-based models are widely used to study dynamical systems in a variety of scientific and engineering problems. However, these models are necessarily approximations of reality due to incomplete knowledge or excessive complexity in modeling underlying processes. As a result, they often produce biased simulations due to inaccurate parameterizations or approximations used to represent the true physics. In this paper, we aim to build a new physics-guided machine learning framework to monitor dynamical systems. The idea is to use advanced machine learning model to extract complex spatio-temporal data patterns while also incorporating general scientific knowledge embodied in simulated data generated by the physics-based model. To handle the bias in simulated data caused by imperfect parameterization, we propose to extract general physical relations jointly from multiple sets of simulations generated by a physics-based model under different physical parameters. In particular, we develop a spatio-temporal network architecture that uses its gating variables to capture the variation of physical parameters. We initialize this model using a pre-training strategy that helps discover common physical patterns shared by different sets of simulated data. Then, we fine-tune it combining limited observations and adequate simulations. By leveraging the complementary strength of machine learning and domain knowledge, our method has been shown to produce accurate predictions, use less training samples and generalize to out-of-sample scenarios. We further show that the method can provide insights about the variation of physical parameters over space and time in two domain applications: predicting temperature in streams and predicting temperature in lakes.