Constructing low-dimensional ordinary differential equations from chaotic time series of high- or infinite-dimensional systems using radial-function-based regression.

In our previous study [N. Tsutsumi, K. Nakai, and Y. Saiki, Chaos 32, 091101 (2022)1054-150010.1063/5.0100166] we proposed a method of constructing a system of ordinary differential equations of chaotic behavior only from observable deterministic time series, which we will call the radial-function-based regression (RfR) method. The RfR method employs a regression using Gaussian radial basis functions together with polynomial terms to facilitate the robust modeling of chaotic behavior. In this paper, we apply the RfR method to several example time series of high- or infinite-dimensional deterministic systems, and we construct a system of relatively low-dimensional ordinary differential equations with a large number of terms. The examples include time series generated from a partial differential equation, a delay differential equation, a turbulence model, and intermittent dynamics. The case when the observation includes noise is also tested. We have effectively constructed a system of differential equations for each of these examples, which is assessed from the point of view of time series forecast, reconstruction of invariant sets, and invariant densities. We find that in some of the models, an appropriate trajectory is realized on the chaotic saddle and is identified by the stagger-and-step method.

Characterizing Small-Scale Dynamics of Navier-Stokes Turbulence with Transverse Lyapunov Exponents: A Data Assimilation Approach.

Data assimilation (DA) of turbulence, which involves reconstructing small-scale turbulent structures based on observational data from large-scale ones, is crucial not only for practical forecasting but also for gaining a deeper understanding of turbulent dynamics. We propose a theoretical framework for DA of turbulence based on the transverse Lyapunov exponents (TLEs) in synchronization theory. Through stability analysis using TLEs, we identify a critical length scale as a key condition for DA; turbulent dynamics smaller than this scale are synchronized with larger-scale turbulent dynamics. Furthermore, considering recent findings for the maximal Lyapunov exponent and its relation with the TLEs, we clarify the Reynolds number dependence of the critical length scale.

Constructing differential equations using only a scalar time-series about continuous time chaotic dynamics.

We propose a simple method of constructing a system of differential equations of chaotic behavior based on the regression only from scalar observable time-series data. The estimated system enables us to reconstruct invariant sets and statistical properties as well as to infer short time-series. Our successful modeling relies on the introduction of a set of Gaussian radial basis functions to capture local structures. The proposed method is used to construct a system of ordinary differential equations whose orbit reconstructs a time-series of a variable of the well-known Lorenz system as a simple but typical example. A system for a macroscopic fluid variable is also constructed.

Dynamical system analysis of a data-driven model constructed by reservoir computing.

This study evaluates data-driven models from a dynamical system perspective, such as unstable fixed points, periodic orbits, chaotic saddle, Lyapunov exponents, manifold structures, and statistical values. We find that these dynamical characteristics can be reconstructed much more precisely by a data-driven model than by computing directly from training data. With this idea, we predict the laminar lasting time distribution of a particular macroscopic variable of chaotic fluid flow, which cannot be calculated from a direct numerical simulation of the Navier-Stokes equation because of its high computational cost.

Low-dimensional paradigms for high-dimensional hetero-chaos.

The dynamics on a chaotic attractor can be quite heterogeneous, being much more unstable in some regions than others. Some regions of a chaotic attractor can be expanding in more dimensions than other regions. Imagine a situation where two such regions and each contains trajectories that stay in the region for all time-while typical trajectories wander throughout the attractor. Furthermore, if arbitrarily close to each point of the attractor there are points on periodic orbits that have different unstable dimensions, then we say such an attractor is "hetero-chaotic" (i.e., it has heterogeneous chaos). This is hard to picture but we believe that most physical systems possessing a high-dimensional attractor are of this type. We have created simplified models with that behavior to give insight into real high-dimensional phenomena.

Machine-learning inference of fluid variables from data using reservoir computing.

We infer both microscopic and macroscopic behaviors of a three-dimensional chaotic fluid flow using reservoir computing. In our procedure of the inference, we assume no prior knowledge of a physical process of a fluid flow except that its behavior is complex but deterministic. We present two ways of inference of the complex behavior: the first, called partial inference, requires continued knowledge of partial time-series data during the inference as well as past time-series data, while the second, called full inference, requires only past time-series data as training data. For the first case, we are able to infer long-time motion of microscopic fluid variables. For the second case, we show that the reservoir dynamics constructed from only past data of energy functions can infer the future behavior of energy functions and reproduce the energy spectrum. It is also shown that we can infer time-series data from only one measurement by using the delay coordinates. This implies that the obtained reservoir systems constructed without the knowledge of microscopic data are equivalent to the dynamical systems describing the macroscopic behavior of energy functions.

Network analysis of chaotic systems through unstable periodic orbits.

A chaotic motion can be considered an irregular transition process near unstable periodic orbits embedded densely in a chaotic set. Therefore, unstable periodic orbits have been used to characterize properties of chaos. Statistical quantities of chaos such as natural measures and fractal dimensions can be determined in terms of unstable periodic orbits. Unstable periodic orbits that can provide good approximations to averaged quantities of chaos or turbulence are also known to exist. However, it is not clear what type of unstable periodic orbits can capture them. In this paper, a model for an irregular transition process of a chaotic motion among unstable periodic orbits as nodes is constructed by using a network. We show that unstable periodic orbits which have lots of links in the network tend to capture time averaged properties of chaos. A scale-free property of the degree distribution is also observed.

Reconstruction of chaotic saddles by classification of unstable periodic orbits: Kuramoto-Sivashinsky equation.

The unstable periodic orbits (UPOs) embedded in a chaotic attractor after an attractor merging crisis (MC) are classified into three subsets, and employed to reconstruct chaotic saddles in the Kuramoto-Sivashinsky equation. It is shown that in the post-MC regime, the two chaotic saddles evolved from the two coexisting chaotic attractors before crisis can be reconstructed from the UPOs embedded in the pre-MC chaotic attractors. The reconstruction also involves the detection of the mediating UPO responsible for the crisis, and the UPOs created after crisis that fill the gap regions of the chaotic saddles. We show that the gap UPOs originate from saddle-node, period-doubling, and pitchfork bifurcations inside the periodic windows in the post-MC chaotic region of the bifurcation diagram. The chaotic attractor in the post-MC regime is found to be the closure of gap UPOs.

Manifold structures of unstable periodic orbits and the appearance of periodic windows in chaotic systems.

Manifold structures of the Lorenz system, the Hénon map, and the Kuramoto-Sivashinsky system are investigated in terms of unstable periodic orbits embedded in the attractors. Especially, changes of manifold structures are focused on when some parameters are varied. The angle between a stable manifold and an unstable manifold (manifold angle) at every sample point along an unstable periodic orbit is measured using the covariant Lyapunov vectors. It is found that the angle characterizes the parameter at which the periodic window corresponding to the unstable periodic orbit finishes, that is, a saddle-node bifurcation point. In particular, when the minimum value of the manifold angle along an unstable periodic orbit at a parameter is small (large), the corresponding periodic window exists near (away from) the parameter. It is concluded that the window sequence in a parameter space can be predicted from the manifold angles of unstable periodic orbits at some parameter. The fact is important because the local information in a parameter space characterizes the global information in it. This approach helps us find periodic windows including very small ones.

Amplitude-phase synchronization at the onset of permanent spatiotemporal chaos.

Amplitude and phase synchronization due to multiscale interactions in chaotic saddles at the onset of permanent spatiotemporal chaos is analyzed using the Fourier-Lyapunov representation. By computing the power-phase spectral entropy and the time-averaged power-phase spectra, we show that the laminar (bursty) states in the on-off spatiotemporal intermittency correspond, respectively, to the nonattracting coherent structures with higher (lower) degrees of amplitude-phase synchronization across spatial scales.

Time-averaged properties of unstable periodic orbits and chaotic orbits in ordinary differential equation systems.

It has recently been found in some dynamical systems in fluid dynamics that only a few unstable periodic orbits (UPOs) with low periods can give good approximations to the mean properties of turbulent (chaotic) solutions. By employing three chaotic systems described by ordinary differential equations, we compare time-averaged properties of a set of UPOs and those of a set of segments of chaotic orbits. For every chaotic system we study, the distributions of a time average of a dynamical variable along UPOs with lower and higher periods are similar to each other and the variance of the distribution is small, in contrast with that along chaotic segments. The distribution seems to converge to some limiting distribution with nonzero variance as the period of the UPO increases, although that along chaotic orbits inclines to converge to a delta -like distribution. These properties seem to lie in the background of why only a few UPOs with low periods can give good mean statistical properties in dynamical systems in fluid dynamics.