What makes a steady flow to favour kinematic magnetic field generation: A statistical analysis

TL;DR

This study addresses why certain steady conducting flows are more effective at magnetic-field generation in the kinematic dynamo regime. It constructs a large ensemble of $2193$ steady 2.5D flows, solves the linear magnetic induction equation to obtain the dominant eigenvalue $\lambda_d$ at $\eta=0.03$, and tests correlations with hydrodynamic metrics such as $\| abla\times{\bf v}\|$ and $\langle{\bf v}\cdot(\nabla\times{\bf v})\rangle$, finding no strong predictive link. The key finding is that simple hydrodynamic quantities do not correlate with dynamo growth, suggesting the need for data-driven methods (e.g., convolutional neural networks) to learn from flow fields and identify features tied to magnetic-field amplification. The authors propose using CNNs and Grad-CAM to extract physically meaningful regions and descriptors, aiming to inform dynamo theory, experiments, and geophysical/engineering MHD applications.

Abstract

To advance our understanding of the magnetohydrodynamic (MHD) processes in liquid metals, in this paper we propose an approach combining the classical methods in the dynamo theory based on numerical simulations of the partial differential equations governing the evolution of the magnetic field with the statistical methods. In this study, we intend to answer the following ``optimization'' question: Can we find a statistical explanation what makes a flow to favour magnetic field generation in the linear regime (i.e. the kinematic dynamo is considered), where the Lorenz force is neglected? The flow is assumed to be steady and incompressible, and the magnetic field generation is governed by the magnetic induction equation. The behaviour of its solution is determined by the dominant (i.e. with the largest real part) eigenvalue of the magnetic induction operator. Considering an ensemble of 2193 randomly generated flows, we solved the kinematic dynamo problem and performed an attempt to find a correlation between the dominant eigenvalue and the standard quantities used in hydrodynamics -- vorticity and kinetic helicity. We have found that there is no visible relation between the property of the flow to be a kinematic dynamo and these quantities. This enables us to conclude that the problem requires a more elaborated approach to ``recognize'' if the flow is a dynamo or not; we plan to solve it using contemporary data-driven approach based on deep neural networks.

Figures (3)

• Figure 1: Dominant eigenvalues, $\lambda_d\,,$ of the magnetic induction operator. The vertical red line separates non-dynamos (left half-plane) from dynamos (right half-plane).
• Figure 2: Histogram of the real and imaginary parts of the dominant eigenvalues of the magnetic induction operator. For complex conjugated pairs only the eigenvalue with non-negative imaginary part is used.
• Figure 3: Contour plots of the kinetic energy density (first column), vorticity (second column) and kinetic helicity (third column) for the flows, which are least (first two lines) and most (last two lines) favourable to magnetic field generation. The growth rates of the magnetic fields are -0.072 (first line), -0.069 (second), 0.084 (third) and 0.088 (fourth).