Investigation of dissipative particle dynamics with colored noise

Morgane Borreguero, Marco Ellero, Nikolaus A. AdamsarXiv 2019 · DOI · arXiv

Abstract

We investigate the behavior of dissipative particle dynamics (DPD) with time-correlated random noise. A new stochastic force for DPD is proposed which consists of a random force whose noise has an algebraic correlation proportional to 1/t and is generated by the so called Kangaroo process. We show that the system exhibits significantly different properties from classical DPD, driven by Wiener noise. While the probability distribution function of the velocity is Gaussian, the acceleration develops a bi-modal character. Although the fluctuation dissipation theorem may not strictly hold, we demonstrate that the system reaches equilibrium states with fluctuation-dissipation balance.

Introduction

Dissipative Particle Dynamics (DPD) is a widely used mesoscopic simulation technique for fluid systems. Traditionally, DPD assumes a Markovian framework where random forces are modeled as "white noise"—uncorrelated in time. However, many complex physical systems, such as biological cells or diffusion in crystal lattices, exhibit "memory effects" where the current state depends on past interactions. To capture these non-Markovian dynamics without the computational complexity of explicit memory kernels, this study explores the use of "colored noise."

Methodology: The Kangaroo Process

The researchers introduced Colored-noise DPD (C-DPD), replacing standard Wiener process noise with the Kangaroo process (KP). The KP generates a stepwise random function that remains constant for a random interval before jumping to a new value. This results in an algebraic correlation ($1/t$), effectively embedding long-term memory into the stochastic force.

Because the standard Fluctuation-Dissipation Theorem (FDT) does not strictly apply to this non-Markovian setup, the authors used empirical calibration to find the balance between friction ($\gamma$) and noise amplitude ($\alpha$). They established a power-law relationship, $\alpha = 5.23\gamma^{0.89}$, to maintain thermal equilibrium.

Key Insights: Bi-modal Acceleration

One of the most significant findings is the divergence in probability distributions for velocity and acceleration. While particle velocity remains Gaussian (following a standard bell curve), the acceleration develops a distinct bi-modal character.

Bi-modal distribution of particle acceleration in C-DPD.

This bi-modality suggests the system oscillates between two most likely acceleration states, a direct consequence of the stepwise nature of the Kangaroo process. This behavior leads to "free flight" trajectories, where particles maintain constant acceleration for short durations before sudden directional changes, resembling Levy flights.

Results and Thermal Stability

Despite the unconventional noise source, the C-DPD system successfully reaches unique stationary equilibrium states. Simulations showed that improper balancing of noise and friction leads to heating or cooling, but the calibrated parameters ensure a stable target temperature.

Thermal stability and relaxation to equilibrium in C-DPD compared to standard DPD.

Furthermore, the Velocity Auto-Correlation Function (VACF) in C-DPD exhibits algebraic decay at long time scales, a hallmark of complex fluids that contrasts with the exponential decay seen in standard DPD. Crucially, the relationship between diffusion and friction remains consistent with standard models, suggesting C-DPD can be integrated into existing simulation workflows.

Conclusion

C-DPD offers a simplified yet powerful approach to modeling memory effects in mesoscopic systems. By altering the nature of the noise itself, researchers can capture complex non-Markovian dynamics and unique acceleration profiles while maintaining thermal stability. Future work will focus on re-introducing conservative forces and exploring local momentum conservation in pairwise interactions.

Summary updated

Abstract

We investigate the behavior of dissipative particle dynamics (DPD) with time-correlated random noise. A new stochastic force for DPD is proposed which consists of a random force whose noise has an algebraic correlation proportional to 1/t and is generated by the so called Kangaroo process. We stress the benefits of a time correlated noise in stochastic systems. We show that the system exhibits significantly different properties from classical DPD, driven by Wiener noise. While the probability distribution function of the velocity is Gaussian, the acceleration develops a bi-modal character. Although the fluctuation dissipation theorem may not strictly hold, we demonstrate that the system reaches equilibrium states with fluctuation-dissipation balance. We believe that our explorative research on the DPD model may stimulate the application of modified DPD to unconventional problems beyond molecular modeling.
BibTeX
@misc{borreguero2019investigation,
  author = {Morgane Borreguero and Marco Ellero and Nikolaus A. Adams},
  title = {{Investigation of dissipative particle dynamics with colored noise}},
  year = {2019},
  eprint = {1901.02451},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1901.02451},
}

Loading related content...

Figure