Magnetic nanoparticles and their magnetization dynamics play an important role in many applications. We focus on magnetization dynamics in large ensembles of single domain nanoparticles being characterized by either Brownian or Néel rotation mechanisms. Simulations of the respective behavior are obtained by solving advection–diffusion equations on the sphere, for which a unified computational framework is developed and investigated. This builds the basis for solving two parameter identification problems, which are formulated in the context of the chosen application, magnetic particle imaging. The functionality of the computational framework is illustrated by numerical results in the parameter identification problems either compared quantitatively or qualitatively to measured data.

磁性纳米粒子及其磁化动力学在许多应用中发挥着重要作用。我们专注于以布朗或尼尔旋转机制为特征的大型单畴纳米粒子集合中的磁化动力学。通过求解球体上的对流-扩散方程来获得各自行为的模拟，为此开发和研究了统一的计算框架。这为解决两个参数识别问题奠定了基础，这些问题是在所选应用（磁性粒子成像）的背景下制定的。计算框架的功能通过参数识别问题中的数值结果来说明，无论是定量还是定性地与测量数据进行比较。