Numerically efficient modelling and simulation of integrated ac-dc power systems using dynamic phasor type solutions

Modern power systems, from continent-spanning networks down to isolated microgrids, are experiencing unprecedented technological changes with broader use of direct current (dc) in addition to traditional alternating current (ac). Such integrated ac-dc power systems present notable challenges in all aspects of design, analysis, control, and operation, where extensive computer simulations play the essential and enabling role. Due to the use of diverse types of signal representation and component formulation, state-of-the-art power system simulation tools are limited to their distinct time scales of transient phenomena. This thesis considers the dynamic phasor (DP) type modelling approaches, where two types of DP theories, namely the shifted-frequency analysis (SFA) and the generalized averaging method (GAM), are considered. In DP-type simulations, power systems are modelled using analogous low-pass time-phasor signals, thereby offering flexible selection of time-step sizes and superior combination of numerical accuracy and efficiency. The ultimate goal of this research is to increase the numerical efficiency of DP-type simulations for the integrated ac-dc power systems. This is achieved by proposing several new DP component models with desirable features and improved numerical properties. First, the constant-parameter SFA model of synchronous machines is proposed to avoid numerically-costly recalculations of the time-varying stator-network matrices. This model is then extended to induction machines for modelling in state-variable based (SV-based) simulation tools. Next, we propose a new, highly efficient model of line-commutated rectifiers using a parametric DP formulation, which is demonstrated as valid for various system operating conditions. Moreover, the effect of ac side harmonics is incorporated to improve modelling fidelity. Finally, the interface between SFA- and GAM-type DPs is achieved to interconnect the proposed DP models. Rigorous case studies demonstrate the superior numerical efficiency of the proposed models, and their advantageous accuracy in capturing the desired phenomena of ac-dc power systems. It is envisioned that the proposed models will become highly useful to many researchers and engineers worldwide, and facilitate the development of next-generation power system simulation tools.Applied Science, Faculty ofElectrical and Computer Engineering, Department ofGraduat