Models, Methods, and Tools to estimate the Power Quality of Electric Systems in the presence of Electric Vehicles

The rapid deployment of Electric Vehicle Charging Stations (EVCS) poses significant technical challenges for modern Distribution Systems (DS), particularly regarding grid integration and power quality degradation. An EVCS at the grid interface is a complex power-electronic load that can significantly impact voltage profiles, line currents, transformer loading, and network losses. In addition to the low-order harmonics, the high-frequency switching phenomena of active power converters in modern charging systems generate significant spectral emissions in the supraharmonic range (2 to 150 kHz). Therefore, it is essential to have strong analytical models, methodologies, and computational tools to accurately estimate waveform distortion and the hosting capacity of DS under realistic operating conditions. This doctoral thesis addresses these issues by introducing a unified assessment tool that encompasses deterministic modeling, probabilistic analysis, and advanced methodologies for hosting capacity assessment. The thesis is broken down into three major thematic blocks, consistent with the progression of this research's structure: First, a deterministic approach is developed to evaluate the initial harmonic and supraharmonic emissions from EVCS (Chapter 2). A low-power single-phase charger is modeled in the time domain to achieve an optimal compromise between mathematical accuracy and computational efficiency: a simplified yet highly accurate model based on a unidirectional power-factor correction (PFC) topology. Deterministic analyses are performed to estimate the effects of changing grid impedance parameters and charging power levels, and to explicitly quantify the effects of Electromagnetic Interference (EMI) filters. The results show that EMI filters have very little effect on structural emissions. However, the Total Harmonic Distortion of Current (THDC) decreases as power increases, whereas the Total Supra-Harmonic Current (TSHC) remains relatively stable with respect to changes in power. This deterministic model is then extended to multiple EVCS configurations, and new Aggregation Factors (AF) and Diversity Factors (DF) are defined to mathematically describe the attenuation and the summation trends of emissions in different frequency bands. Second, a probabilistic assessment framework is formulated to account for the stochastic nature of real-world charging behavior (Chapter 3). The structural uncertainties associated with EVCS grid integration are addressed using advanced simulation techniques, such as Monte Carlo (MC) and Quasi-Monte Carlo (QMC) simulation methods. Four different probability density functions (Gaussian, Uniform, Beta Right, and Beta Left) are used to model the stochastic variability of charging power. The mean and variance of the low-order THDC are highly sensitive to the distribution of charging power, whereas the TSHC profiles appear to be structurally invariant with respect to it. Moreover, the findings indicate that high-frequency supraharmonic emissions are more robust to stochastic power changes than low-harmonic emissions in the case of multiple EVCS deployments. Third, the dissertation explores how to quantify Electric Vehicle Hosting Capacity (EVHC) in distribution networks (Chapter 4). An iterative load-flow approach is developed to systematically analyze the EVHC under strict operational constraints, considering four boundary conditions: baseline network, transformer thermal limits, line current ampacity limits, and nodal voltage deviation thresholds. The iterative approach is validated on a modified IEEE 69-bus system and expanded to include the simultaneous effects of Distributed Generation (DG) and the systemic effects of supraharmonics. Finally, a physically grounded and scalable evaluation tool, the Capability Chart model, is introduced. This framework also directly measures the maximum EVCS load capacity that can be assigned without violating grid constraints by mapping the admissible operational boundaries onto the IEEE 33-bus system's active-reactive (P-Q) power plane, yielding a comprehensive, mathematically sound tool for modern DS planning and operation.