Remarkable evolutionary laws of absolute and relative entropies with dynamical systems

TL;DR

This paper derives fundamental laws governing the evolution of absolute and relative entropies in dynamical systems, revealing that relative entropy remains conserved while absolute entropy can evolve under specific conditions, with implications for chaos analysis.

Contribution

It establishes new laws for entropy evolution in dynamical systems, showing relative entropy conservation and conditions for absolute entropy change, enhancing understanding of chaos and ensemble predictions.

Findings

Relative entropy $D$ is conserved in deterministic systems.

Absolute entropy $H$ evolves under additive noise and nondivergent flows.

Lyapunov exponent for the density function is always zero, indicating stability in the phase space.

Abstract

The evolution of entropy is derived with respect to dynamical systems. For a stochastic system, its relative entropy $D$ evolves in accordance with the second law of thermodynamics; its absolute entropy $H$ may also be so, provided that the stochastic perturbation is additive and the flow of the vector field is nondivergent. For a deterministic system, $dH/dt$ is equal to the mathematical expectation of the divergence of the flow (a result obtained before), and, remarkably, $dD/dt = 0$. That is to say, relative entropy is always conserved. So, for a nonlinear system, though the trajectories of the state variables, say $\ve x$, may appear chaotic in the phase space, say $\Omega$, those of the density function $\rho(\ve x)$ in the new ``phase space'' $L^1(\Omega)$ are not; the corresponding Lyapunov exponent is always zero. This result is expected to have important implications for the ensemble predictions in many applied fields, and may help to analyze chaotic data sets.