Today, the electric vehicle (EV) is the best positioned option to massively replace traditional combustion cars in the market. However, despite the optimistic forecasts, there are still many factors that prevent its deployment; the current global VE stock corresponds to only 0.2% of the total number of passenger vehicles in circulation. As a consequence, the automotive electric drives of future vehicles have to overcome a number of technological challenges in order improve its efficiency. High power density EV power inverters require a new approach to cope with stringent requirements of high current density. The heat flux of power modules for EVs is high, and it is projected to increase as the current densities and switching frequencies increase, so new efficient cooling technologies are necessary. The reliability of power electronic systems is closely dependent on the thermal behavior of semiconductor devices. Many technologies and techniques for cooling EV power modules can be found in the literature. At the beginning of this thesis, a review and classification of the main thermal management techniques is presented. Furthermore, in order to develop a power module for EV applications new solutions for module integration and packaging technology are needed. In this sense, the technical trends and advances in EV power module packaging technologies will be reviewed. On the other hand, several papers in the scientific literature present results on two-phase cooling techniques for the removal of high heat flux. However, no direct experimental data is available for comparison of single-phase and two-phase cooling of EV power inverter working through the same full drive cycle, in order to assess both systems. The main goal of this thesis is to test and compare an experimental VCTPL (Vapour Compression Two-Phase Loop) prototype built using conventional automotive air conditioning components (condenser, expansion valve, compressor, and vapour and liquid lines) with a conventional single-phase cooling system under the same drive cycle. Because the heat transfer coefficient, h, is not a parameter provided by manufacturers of cold plates for conventional single-phase cooling, will be necessary to determine experimentally the value of h for a specific cold plate, before detailed three-dimensional FEM (Finite Element Method) simulation of the two-phase cooling in the motor inverter will be done. In this way, simulation of the complex two-phase phenomenon, with thermal transport by boiling and convection, will be simplified by substitution of all the mechanisms by a simple uniform heat transfer coefficient, h, along the inner walls of the cold plate.