Hosting capacity calculations are used to quantify the limitations of electrical networks regarding the installation of new production or consumption. The concept is based on evaluating the performance of the network with increasing amounts of distributed energy resources. Since it involves planning for future conditions of a highly complex system, the calculations are subject to multiple uncertainties, which also extend to the results. Having a clear understanding of how these uncertainties are modelled is critical to a correct interpretation of the results from hosting capacity studies. 

This thesis aims to improve the understanding of different aspects within hosting capacity calculation methods and propose different modelling approaches including low and medium voltage levels, and their interdependence. 

The wide range of options within the hosting capacity calculation methods and the lack of standardisation and defined terminology make it easy to overlook some of the basic principles. This work offers an overview of different hosting capacity calculation methods, accounting for different modelling approaches, the associated uncertainties, and how they affect the interpretation of the results. These insights encourage a reconnection with the fundamental principles that guide the concept of hosting capacity, facilitating a more informed analysis of the results. 

The first part of the work applies a hosting capacity calculation method that considers all possible combinations of locations for PV installations, and based on that, calculates the probability of a limit violation happening. The probability of overvoltage or overloading is used as the performance index and a planning risk is needed to define the limit of acceptable performance. 

The background voltage refers to the voltage magnitude before the installation of any new production or consumption. It accounts for the impact on the voltage of production and consumption in other parts of the distribution network. This thesis introduces a time-dependent model for the background voltage in order to assess the hosting capacity of a LV network, including the impact of already existing PV installations in other LV networks fed by the same MV feeder. 

The background voltage model was extended to analyse the impact of MV reserve operating paths on the hosting capacity of the LV networks. It represents a temporary MV configuration, and the analysis includes the impact of consumption and production throughout the year and along the MV feeder. The analyses results in the impact on the hosting capacity, as well as how long and how many customers it would impact with different reserve operating paths. 

The second part of the work investigates the impact of bigger installations at the MV level. A visualisation method is proposed to assess the trade-off between new installations at MV and LV. This approach enables the evaluation of how installations at one voltage level can constrain the available capacity on the other. The method supports the estimation of a global hosting capacity to be shared among customers across voltage levels. 

In the third part of the work, a model for grid strength for active and reactive power is presented. This model is used to assess the hosting capacity for multiple customers and evaluate the impact of reactive power compensation. 

The findings from this thesis cover multiple aspects of hosting capacity calculations, reveal the interdependency between voltage levels and provide methods to address and visualise the mutual dependency between new installations at LV and MV networks. A critical analysis and comprehensive discussion of different ways the hosting capacity can be modelled and shared between customers is presented, as well as recommendations for future work.