Abstract
In the study of dynamical systems, the transition from autonomous to non-autonomous models represents a significant increase in complexity. While classical dynamics focuses on the repeated application of a single map, non-autonomous discrete dynamical systems (NADDS) involve a sequence of maps that change over time. This paper, originally published in the Journal of the Korean Mathematical Society, formalizes the concepts of $\omega$-limit sets and attraction within this time-varying context. The authors demonstrate that while certain classical properties—such as set invariance—are lost, the fundamental relationship between attraction and limit sets remains robust in regular topological spaces.
Introduction
A standard autonomous discrete dynamical system is defined by a space $X$ and a single function $f$, where one studies the sequence of iterates $x, f(x), f(f(x)), \dots$. However, many real-world phenomena, including climate patterns and economic fluctuations, are better modeled as non-autonomous systems where the governing rules evolve. These are represented by a sequence of maps $(f_1, f_2, f_3, \dots)$, where the state at time $n$ is determined by the composition of the preceding $n$ functions.
The primary challenge in NADDS is that the mathematical tools used to describe "eventual" behavior—specifically where an orbit accumulates in the limit—become significantly more complex. This research addresses the theoretical gap in defining and connecting $\omega$-limit sets (the set of points an orbit accumulates around) and attraction (the tendency of orbits to converge toward a specific set) for these non-constant sequences.
Methodological Approach
The research is grounded in point-set topology and dynamical systems theory. To explore the properties of NADDS, the authors employ several key mathematical strategies:
- Formal Definition Extension: The classical definitions of orbits and limit sets are adapted to accommodate a sequence of distinct maps $(f_n)_{n=1}^\infty$.
- Topological Analysis: The study investigates these properties within the context of "regular spaces" (topological spaces where points and closed sets can be separated by neighborhoods) and compact spaces.
- Counterexample Construction: The authors use specific sequences of maps to illustrate where non-autonomous systems deviate from classical expectations, providing a boundary for theoretical generalizations.
- Topological Semi-conjugacy: This technique is used to demonstrate that attraction is a robust property that persists even when a system is transformed by a continuous, surjective map.
Theoretical Contributions
The Loss of Invariance
In classical autonomous dynamics, $\omega$-limit sets are invariant; if a point is in the set, its image under the map $f$ is also in the set. A key finding of this paper is that this property does not hold for non-autonomous systems. Because the maps change at each step, an $\omega$-limit set can "drift," and the sequence of maps does not necessarily map the set back into itself. This highlights a fundamental divergence from traditional dynamical theory.
Robustness of Attraction
The authors prove that attraction is topologically robust. Specifically, if one system attracts a set of points, any system that is "semi-conjugate" to it—essentially a simplified or re-scaled version of the system—will also exhibit attraction. This provides a mathematical justification for using simplified models to study complex, time-varying systems.
The Attraction-Limit Set Connection
The most significant theoretical contribution is found in Theorem 4.2. The authors prove that in a regular space, if a compact set $K$ attracts a set $E$, then the $\omega$-limit set of $E$ is not merely a random collection of points. Instead, it is a non-empty, compact set that itself attracts $E$. This result bridges the geometric notion of attraction with the set-theoretic notion of limit points, ensuring consistency in how we define long-term behavior.
Core Findings and Results
- Set Operations: The research establishes that the $\omega$-limit set of a union of two sets is the union of their individual $\omega$-limit sets: $\omega(A \cup B) = \omega(A) \cup \omega(B)$.
- Periodic Systems: For $k$-periodic systems (where the sequence of maps repeats every $k$ steps), the authors show that non-autonomous attraction properties can be reduced to the study of an "induced" autonomous system formed by the composition of the $k$ maps.
- Existence Criteria: The paper provides sufficient conditions under which an $\omega$-limit set is guaranteed to be non-empty and compact, which is vital for ensuring that the concept of "long-term behavior" is well-defined in a given system.
Limitations and Future Directions
While providing a rigorous foundation, the study's results are primarily constrained to regular spaces. The authors do not explore how these properties might degrade in more pathological or non-Hausdorff topological spaces. Additionally, many proofs rely on the assumption of compactness, which may limit direct application in infinite-dimensional spaces or non-compact settings common in certain physical applications.
Future research may build upon this groundwork to explore non-autonomous chaos, topological entropy, and "mixing" properties. A natural extension of this work would be investigating how these $\omega$-limit sets behave under random perturbations rather than deterministic sequences, moving closer to modeling real-world stochasticity.
Summary updated