Physics-Guided Deep Learning for Drag Force Prediction in Dense Fluid-Particulate Systems

Abstract

Physics-based simulations are often used to model and understand complex physical systems in domains such as fluid dynamics. Such simulations, although used frequently, often suffer from inaccurate or incomplete representations either due to their high computational costs or due to lack of complete physical knowledge of the system. In such situations, it is useful to employ machine learning (ML) to fill the gap by learning a model of the complex physical process directly from simulation data. However, as data generation through simulations is costly, we need to develop models being cognizant of data paucity issues. In such scenarios, it is helpful if the rich physical knowledge of the application domain is incorporated in the architectural design of ML models. We can also use information from physics-based simulations to guide the learning process using aggregate supervision to favorably constrain the learning process. In this article, we propose PhyNet, a deep learning model using physics-guided structural priors and physics-guided aggregate supervision for modeling the drag forces acting on each particle in a computational fluid dynamics-discrete element method. We conduct extensive experiments in the context of drag force prediction and showcase the usefulness of including physics knowledge in our deep learning formulation. PhyNet has been compared with several state-of-the-art models and achieves a significant performance improvement of 7.09% on average. The source code has been made available*.

Physics Guided Deep Learning for Data-Driven Aircraft Fuel Consumption Modeling

“…We can consider the scalar λ as a ratio between a constant scalarλ, and pressure ratio multiplied by square root of temperature ratio. This assumption is based on Equation (17).…”

“…In order to obtain more accurate results and reliable out-of-sample generalization, the vital intention is to merge physics-based models with ML algorithms to leverage their complementary capabilities. Such combined ML-physics models are expected to thoroughly seize the dynamics of scientific systems and improve the knowledge of underlying physical laws [17]. There are several ways to inject physical laws, knowledge, or information into ML models to build physics aware ML models [18].…”

Multi-End Physics-Informed Deep Learning for Seismic Response Estimation

“…Inspired by this, PIDL has been extended to engineering applications. The related studies have involved many communities, including fluid dynamics [ 15 , 23 , 24 ], geology [ 25 ], fatigue analysis [ 26 ], power system [ 27 ], and system identification [ 28 ] and controls [ 29 ]. In PIDL, a conventional strategy is writing the governing PDE into the loss function [ 30 ] to compress the solution space.…”

Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning

“…Thus, there is a need for a new kind of AI models that are more efficient regarding safety, interpretability, and explainability, with a promising viable solution in this direction being represented by the use of so-called Informed ML [5] approaches where AI models can be improved by using additional prior knowledge into their learning process. Recently, this approach is proving to be successful in many fields and applications such as lake temperature modeling [4], MRI reconstruction [6], real-time irrigation management [7], structural health monitoring [8], fusion plasmas [9], fluid dynamics [10] and machining tool wear prediction [11]. However, regarding autonomous driving, this approach was not fully explored, with recent research projects such as KI Wissen [12] funded by the German Federal Ministry for Economic Affairs and Energy being one of of the first, if not, the first one that tries to bring knowledge integration into Automotive AI in order to increase their safety.…”