The variable electricity output of the RES (renewable energy sources) power plants, such as wind and hydropower, is an important challenge for the electricity system managers. This paper addresses the problem of an electricity system supported mainly on hydro, thermal, and wind power plants. A binary mixed integer non-linear optimization model with hourly time step is described. The model is applied to a system close to the Portuguese electricity case assuming demand forecasts for the year 2020. The main objective of this paper was to analyze the impact that different levels of installed wind power can have in the operation of this electricity system, taking into account the hourly and intra-annual variation of the renewable resources, the demand projections and also the technical restriction of thermal power plants. The results confirmed wind power as strategic technology to reduce both the marginal cost and CO2 emissions. According to the simulations run, wind power will not replace hydropower but a decrease of thermal power production is foreseen as more wind power is added to the system. Large wind power scenarios will particularly affect gas power plants performance, reducing both the load level and the number of operating hours. (C) 2014 Elsevier Ltd. All rights reserved.