Migraine is a highly disabling and complex neurological disorder with global impact. The relatively complex mechanism of migraine prediction and regulation has become a basic scientific problem restricting the standardization of clinical diagnosis and treatment of migraine. Related experimental data shows that the process of migraine exhibits significant nonlinear fluctuation behavior on multiple spatiotemporal scales. However, research literature on the dynamics of migraines is very limited and lacks a unified theoretical framework to understand the mechanisms of migraine. This article reviews the research progress in migraine dynamics modeling, analysis, and control strategies in recent years, with a focus on the neurocircuit dynamics behavior and bifurcation mechanism of cortical spreading depression (CSD) propagation, as well as the neural computing methods for the design of migraine regulation scheme and the evaluation of efficacy. Based on the current research status of migraine theory, this paper looks forward to the research ideas and issues that should be paid attention to in datadriven dynamic modeling analysis, prediction, and regulation strategies.