Nowadays, hydrogen represents a promising alternative to fossil fuels in its use as an energy carrier for internal combustion engines. The absence of carbon in hydrogen is a key factor in reducing greenhouse gas emissions and unburned hydrocarbons. However, hydrogen combustion unfortunately generates nitrogen oxides (NOx), pollutants that result from the oxidation reaction of molecular nitrogen present in the intake air at high temperatures. In this framework, this thesis aims to develop a semi-predictive phenomenological model for estimating NOx emissions in a hydrogen-powered engine. Starting from the anylisis of experimental data, a mathematical calculation model to predict NOx as a function of the main engine parameters that most influence emission variations will be exposed.