Domain-decoupled Physics-informed Neural Networks with Closed-form Gradients for Fast Model Learning of Dynamical Systems

Henrik Krauss,Tim-Lukas Habich,Max Bartholdt,Thomas Seel,Moritz Schappler

2024-08-28

Abstract:Physics-informed neural networks (PINNs) are trained using physical equations and can also incorporate unmodeled effects by learning from data. PINNs for control (PINCs) of dynamical systems are gaining interest due to their prediction speed compared to classical numerical integration methods for nonlinear state-space models, making them suitable for real-time control applications. We introduce the domain-decoupled physics-informed neural network (DD-PINN) to address current limitations of PINC in handling large and complex nonlinear dynamical systems. The time domain is decoupled from the feed-forward neural network to construct an Ansatz function, allowing for calculation of gradients in closed form. This approach significantly reduces training times, especially for large dynamical systems, compared to PINC, which relies on graph-based automatic differentiation. Additionally, the DD-PINN inherently fulfills the initial condition and supports higher-order excitation inputs, simplifying the training process and enabling improved prediction accuracy. Validation on three systems - a nonlinear mass-spring-damper, a five-mass-chain, and a two-link robot - demonstrates that the DD-PINN achieves significantly shorter training times. In cases where the PINC's prediction diverges, the DD-PINN's prediction remains stable and accurate due to higher physics loss reduction or use of a higher-order excitation input. The DD-PINN allows for fast and accurate learning of large dynamical systems previously out of reach for the PINC.

Systems and Control,Machine Learning,Robotics

What problem does this paper attempt to address?

### Problems Addressed by the Paper

This paper aims to address the challenges faced by Physics-Informed Neural Networks (PINNs) when dealing with large-scale complex nonlinear dynamic systems. Specifically, the paper proposes a Domain Decomposition Physics-Informed Neural Network (DD-PINN) to overcome the limitations of existing methods (such as PINNs for Control, i.e., PINC) in the following aspects:

1. **Long training time due to automatic differentiation**: Traditional PINC uses Graph-based Automatic Differentiation, which can lead to very long training times in large-scale or complex systems.
2. **Low accuracy due to constant excitation assumption**: PINC assumes that the excitation signal remains constant within the prediction interval (zero-order hold assumption), which can lead to lower prediction accuracy over long intervals.
3. **Weighting issues of initial condition loss and physical loss**: Requires the use of weighting schemes (such as GradNorm, SoftAdapt, etc.), which need to be chosen empirically and are sensitive to hyperparameters, further increasing training time.

By introducing DD-PINN, the authors address the above issues and demonstrate its effectiveness on three different dynamic systems: a nonlinear spring-damper system, a five-mass chain system, and a two-link robotic system. Experimental results show that DD-PINN can significantly reduce training time and, in some cases, maintain higher prediction stability and accuracy.