This paper proposes a modeling and simulation procedure of power distribution networks (PDN) based on polynomial fitting and bilinear transformation to generate datasets for data-driven control of DCDC converters. A measurement setup is proposed which enables the definition of PDN capturing its broadband and high dynamic range characteristics. In order to quantify the modeling error, a method for measuring the required signals and for efficient simulation is proposed. The modeling and simulation procedure is applied to a measured PDN and the modeling error is discussed.