Abstract
The Lie derivative plays a key role in mathematics and physics. In particular, the Lie derivative discretization scheme has been implemented for a relatively large time step in order to quickly solve the numerical solution of systems. In this paper, an algorithm of the Lie derivative discretization scheme is applied to variable mass systems. Four different types of variable mass systems are employed to study the numerical solutions, and the calculated results are consistent with those obtained by the fourth-order Runge-Kutta method. Computational experiments demonstrate the success of the proposed method on variable mass systems. Moreover, the algorithm of Lie derivative discretization is shown to have superior computational efficiency and larger time steps compared to the Runge-Kutta algorithm.
Introduction
In classical mechanics, many systems are simplified by assuming constant mass. However, real-world engineering applications—such as rockets consuming fuel, maglev spindles, and pipes with moving interfaces—frequently involve variable mass. These systems are governed by nonlinear differential equations with time-varying coefficients, making analytical solutions nearly impossible to obtain.
While numerical methods like the fourth-order Runge-Kutta (RK4) algorithm are widely used, they can be computationally expensive, particularly for long-duration simulations requiring high precision. This paper explores the application of a Lie derivative discretization scheme to address these challenges, aiming for a balance of high accuracy, computational efficiency, and stability at larger time steps.
Methodology
The proposed approach utilizes the Lie derivative, a fundamental concept in differential geometry that describes the change of a geometric object along a vector field.
Lie Expansion and Discretization
The core of the method involves representing the system's governing equations through a Lie expansion. A second-order discretization formula is then applied to this expansion. Unlike standard Taylor series-based methods, Lie-based schemes are inherently better at preserving the topological features of the system, such as stationary points. This ensures that the numerical solution remains physically faithful even under simplification.
Comparative Framework
To validate the method, the authors implemented the Lie scheme across four benchmark models:
- Variable mass monostable nonlinear oscillator.
- Variable mass symmetric bistable nonlinear oscillator.
- Variable mass asymmetric bistable nonlinear oscillator.
- Variable mass linear oscillator.
The performance of the Lie scheme was compared against the industry-standard RK4 method (e.g., MATLAB's ode45) in terms of accuracy and CPU processing time.
Key Results and Discussion
The computational experiments revealed significant advantages for the Lie derivative scheme across all tested oscillators.
Accuracy and Topological Integrity
The displacement responses and phase diagrams generated by the Lie scheme were found to be virtually identical to those from the RK4 method. This confirms that the Lie-based discretization accurately captures the complex dynamics of variable mass systems.
Computational Efficiency
One of the most striking findings was the leap in efficiency. In tests involving the monostable oscillator, the Lie method was 93.7% more efficient than RK4. For the symmetric bistable system, the Lie method completed simulations in approximately 0.12–0.14 seconds, whereas the RK4 method required 1.85–2.07 seconds for the same parameters.
Robustness at Large Time Steps
Traditional solvers often lose accuracy or become unstable if the time step ($h$) is too large. The Lie scheme demonstrated remarkable robustness, maintaining high accuracy at $h = 0.1$, while standard methods typically require $h = 0.01$ or smaller to maintain the same level of detail.
Conclusion
The application of the Lie derivative discretization scheme to variable mass systems offers a powerful alternative to traditional numerical solvers. By leveraging algebraic pre-calculations and preserving the system's underlying physics, the method achieves superior speed and allows for significantly larger time steps without sacrificing accuracy. Future work may involve extending this to higher-order discretization and testing on complex, real-world aerospace or manufacturing hardware.
Summary updated