The main task of the energy system is to supply the consumers with high-quality electric and heat energy. As possibilities for accumulation of the energy and especially electrical energy in Estonia are very limited, one of the main energy parameters is its uninterrupted supply. The needs of consumers are characterized by the demand curve-the variation of load for a given time period (day, month, year). It is necessary to stress the difference between load and demand curves for the producer and consumer. Up to the recent time the producer load curve consisted of the individual consumers' demand curves sum plus losses in the distribution elements (in electric networks). Nowadays when by economical and ecological reasons the renewable energy sources are more intensively used, the part of the energy producers using wind and solar energy is constantly rising, and they are increasingly influencing the work of the whole energy system. That complicates significantly the work of the high-powered electric energy generators (with large inertia) at power stations as, in addition to the load variations depending on demand, they have to compensate extremely stochastic production of wind turbines. In this paper the problem is discussed on the basis of load and demand curves of the Energy system of Estonia and Pakri wind farm. It is shown that these curves are not suitable for mutual compensation and that may disturb the stability of the energy system at the wind park maximum power. The result is that the Energy system dispatcher is forced to limit the production of the wind park.

One of the most important problems of the Estonian energy system is the compensation of power fluctuations of wind parks. The possibilities of energy system stabilization by Estonian oil shale power plants are limited. At the same time the use of wind energy has not led to the expected reduction in oil shale consumption. Autonomous wind energy systems with some storage devices are one possible way to develop wind energy usage. A system consisting of a wind generator and storage device is analyzed in this article. The storage device might be a fuel cell with hydrogen storage, pumped storage, flywheel or battery. A unit generator and the normalized and averaged power curve of small wind turbine generators are used for the analysis of wind data. Average hourly wind speeds are measured in different locations of Estonia. The duration, frequency and distribution of windless time periods are analyzed. The estimated principles of storage device selection are given according to wind data and power curves.