The European energy market is becoming increasingly complex due to the growing use of renewable, weather-dependent energy sources, the decommissioning of conventional power plants and the increased use of short-term storage facilities. This makes it more difficult to reliably assess resource adequacy, i.e. the ability of power plants to meet electricity demand under various conditions. A Monte Carlo simulation in combination with energy market models is used for the analysis. However, this method is very computationally intensive. To address this problem, the thesis investigated the use of metamodels, which can map the behaviour of the simulation and require significantly less computing power. A single model did not provide sufficient results, but a model pipeline with two machine learning models performed very well. In addition, an active learning approach was developed to specifically select the best training data.