Electrification of the transportation sector is an important part of the transition towards net-zero emissions. The resulting increase in electric-vehicle charging, ranging from residential charging to high-power and megawatt-scale charging, introduces new demands on distribution networks and increases the penetration of power-electronic converters. One consequence is waveform distortion, including harmonics, interharmonics, and supraharmonics, whose magnitude and propagation depend not only on the charger emissions but also on the interaction between the converter and the surrounding network.

This thesis investigates waveform distortion from electric-vehicle charging through a combination of laboratory and field measurements, converter modelling, network analysis, and hosting-capacity studies. Measurements of Mode 3 and Mode 4 charging, together with specialised vehicle-to-vehicle charging cases, show that charger emissions cannot in general be represented by a single fixed spectrum. Harmonic, interharmonic, and supraharmonic emissions vary with charger architecture, operating stage, power level, phase, background distortion, and network impedance. The measurements further show that the nature of the distortion differs across frequency ranges, reinforcing the need for broadband assessment rather than consideration of conventional harmonics alone. 

To represent the interaction between chargers and the network, frequency-coupled admittance models are developed and incorporated into frequency-domain network analysis. The converter response is found to be dominated by same-frequency interactions, while cross-frequency coupling remains relevant, particularly at lower frequencies and around network resonances. The network response is therefore governed jointly by converter admittance, frequency coupling, and frequency-dependent network impedance. The resulting frequency-coupled network equations are solved using an iterative GMRES-based formulation suitable for harmonic and supraharmonic propagation studies. 

The developed models are subsequently used to assess harmonic hosting capacity in distribution networks. The studies show that hosting capacity is not determined by charging power or EV penetration alone, but depends on charger location, phase connection, converter characteristics, population diversity, background distortion, and network impedance. Single-phase charging can become particularly restrictive in the studied low-voltage networks, while heterogeneous charger populations can exhibit different aggregation behaviour from homogeneous worst-case populations. Hosting capacity is therefore formulated not only as an emission margin or deterministic number of chargers, but also probabilistically, allowing the likelihood of violating network limits to be evaluated as EV penetration increases. 

Overall, the thesis demonstrates that EV-related waveform distortion and harmonic hosting capacity should be treated as converter–network interaction problems rather than as properties of the charger alone. The work provides a measurement-based and modelling framework for representing these interactions across harmonic and supraharmonic frequencies and for assessing their consequences for increasing EV penetration in distribution networks