The key driver for fundamental changes in current energy systems is the need to create a sustainable energy future. However, these changes pose major challenges for the energy system. Above all, the electrical grids represent bottlenecks of these developments. They limit the expansion of decentralized, volatile renewable energy sources (RES), as well as modern consumers (electric vehicles (EV), heat pumps (HP), Power-to-Gas (PtG), etc.) due to the required steady balance between generation and consumption to maintain a stable energy supply and to not cause overloads in the electrical equipment. One possibility to make these strict balance requirements in the electrical grid more flexible in the future are multi-energy systems (MES). MES couple currently independently operated energy carrier grids (e.g., electricity, natural gas, and heat) and thus enable intersectoral load shifts to other energy carrier grids. To explore these MES, suitable models are required with which scenarios and simulations can be calculated in advance. Modeling presents as an additional challenge as the individual energy carrier grids are coupled via a higher-level MES framework and therefore have common connection points where similar characteristics of the grids are necessary. Since the individual grid models have different levels of temporal and spatial detail due to physical constraints, modeling these connection points proves difficult. Therefore, this thesis aims to present a comprehensive modeling approach that not only shows a potential approach to define similar grid characteristics for all energy carrier grids, but also provides a compromise between the temporally and spatially highly resoluted electrical grid model to save computation time. This approach includes a network reduction method developed in this thesis, which allows to achieve higher modeling accuracies in the reduced grid models compared to the previous methods and therefore represents the original grid with minimal deviations. For this purpose, the method is validated using customized developed synthetic test grids that represent real grid behavior for all voltage levels. In addition to electrical grid modeling for MES applications, a suitable analysis must be performed to identify the influence of MES on the electrical grid. Therefore, this paper presents three different use cases in which the developed network reduction method is applied to create equivalent reduced electrical grid models. Via subsequently performed hybrid load flow calculations with the multi-energy carrier simulation framework HyFlow, which was developed at the Chair of Energy Network Technology, the effects of hybrid load shifts in MES on the electrical grid are analyzed. For this purpose, voltage quality and stability analyses as well as analyses of thermal line overloads and self-consumption are performed.