During the last 15 years, weather extremes caused several disruptions to the Italian power system. Their increasingly occurrence is mainly due to the exchanges along the meridians of air masses with very different thermal, density and moisture content properties. The Italian transmission system operator and the distribution companies have repeatedly stressed the need to have a reliable and updatable weather dataset with a history of at least 15 years to improve the resilience of the electricity system. The aim of this work is to develop a new MEteorological Reanalysis Italian DAtaset (MERIDA) that is able to respond to power operators, which need reliable meteorological data to implement effective adaptation strategies to operate the electricity system safely. MERIDA consists in a dynamic downscaling of ERA5 using the Weather Research and Forecasting (WRF) model, which is configured to describe typical weather conditions in Italy. Furthermore, the optimal interpolation (OI) technique is applied to modeled 2 m temperature and precipitation data using weather observations from regional agencies for environmental protection. MERIDA is verified against COSMO REA6 of the Deutscher Wetterdienst (DWD) and ERA5 itself for the period 2010-2015, showing comparable or better results in the reconstruction of 2 m temperature and precipitation. The best results are obtained with MERIDA post-processed by the OI. Some severe weather events that determined important power outages are also analyzed, showing that MERIDA can identify the meteorological conditions leading to significant events of wet snow, heatwaves and floods through their correct spatial and temporal location and through a quantitative assessment of each weather event.