Resumen: Pressure relief valves are mechanical devices used in the industry to protect systems against excessive pressure, thus preventing accidents from occurring. Nowadays, a multitude of flows can pass through these pressure relief valves, such as the biphasic ones, in which this master's thesis is going to be framed. These are used in industries such as the oil and chemical industries. The problem comes when trying to characterize them since there is not enough experimental data in the literature to be able to develop a good model, in addition to the fact that if the valves are over-sized or under-sized, it could lead to more economic costs and possible safety problems. In this context, the present Master's thesis aims to find a calculation model that allows the adjustment of some experimental data to predict their behaviour. To achieve this goal, the different methods that can be used for pressure relief valves with two-phase flows will be studied and an analysis of the results will be carried out to find the model that best predicts the behaviour and adjusts to the experimental data provided. In this thesis, we have taken as reference the experimental data provided by the University of La Sapienza, Rome, in which the mass flow rate of a steam/water mixture passing through a valve of 0.01 m diameter has been measured for different initial pressures and qualities of the inlet vapour of the mixture. Therefore, it was decided to try to fit some experimental data to some of the existing models (HEM, HNE-DS, API, ISO) to describe the behaviour of these fluids. The results obtained have shown the differences that exist between the two homogeneous models (HEM, HNE-DS) as well as the similarity of each of them with the other two described (API, ISO). The conclusion has been reached that the model that best fits the behaviour of the data is the homogeneous non-equilibrium model since it is in this model that a greater similarity is found between the experimental mass flow rate provided and the theoretical mass flow rate calculated. This hypothesis has been corroborated thanks to a comparison made to the HEM and HNE-DS models in which it has been concluded that at high pressures the homogeneous equilibrium model is the one that best fits the data but the model has wide spread results, but at the end, for all the experimental data the homogenous non-equilibrium model is the one that in the end best fits the them.