Manifold Learning Approach for Chaos in the Dripping Faucet

Abstract

Dripping water from a faucet is a typical example exhibiting rich nonlinear phenomena. For such a system, the time stamps at which water drops separate from the faucet can be directly observed in real experiments, and the time series of intervals \tau_n between drop separations becomes a subject of analysis. Even if the mass m_n of a drop at the onset of the n-th separation, which cannot be observed directly, exhibits perfectly deterministic dynamics, it sometimes fails to obtain important information from time series of \tau_n. This is because the return plot \tau_n-1 vs. \tau_n may become a multi-valued function, i.e., not a deterministic dynamical system. In this paper, we propose a method to construct a nonlinear coordinate which provides a "surrogate" of the internal state m_n from the time series of \tau_n. Here, a key of the proposed approach is to use ISOMAP, which is a well-known method of manifold learning. We first apply it to the time series of $\tau_n$ generated from the numerical simulation of a phenomenological mass-spring model for the dripping faucet system. It is shown that a clear one-dimensional map is obtained by the proposed approach, whose characteristic quantities such as the Lyapunov exponent, the topological entropy, and the time correlation function coincide with the original dripping faucet system. Furthermore, we also analyze data obtained from real dripping faucet experiments which also provides promising results.