RELEVANCE. Wind power forecasting ensures efficient operation of wind power plants in the electricity markets. Improving the wind power plants output power forecasting accuracy provides more efficient operation planning of the power regions with high installed capacity wind power capacity operation. As a result, the reliability of power supply to these energy regions is increased. PURPOSE. The purpose of this study is to develop an improved power curve approximation technique for fitting the function of the wind power plant output power depending on meteorological factors. The proposed technique should provide a sufficiently low error while maintaining a small computational complexity in the context of a limited amount of retrospective data and the number of independent values (features). METHODS. The MATLAB software package was used to carry out simulation series. Six techniques for obtaining approximation functions were considered during these simulations. Each simulation was carried out in accordance with the following algorithm. First, the parameters of the approximating function were obtained from the training data. Secondly, wind power plant generation forecasts were calculated based on test data. Then the forecasts were compared with the actual values. Finally, the accuracy criteria were estimated. RESULTS. Based on the results of the simulations and comparison of the accuracy evaluation criteria, the most effective technique for the wind farm power curve approximation was identified. A distinguishing feature of this technique is the splitting of the training data into subsamples at 16 wind directions, as well as preprocessing the data in each subsample using the bin method. The proposed approximation technique can be used for short-term and operational forecasting of wind farms output power and electrical energy. CONSLUSTIONS. this paper a number of wind farm power curve approximation techniques were proposed and a comparison of these techniques was carried out. По результатам сравнения рассмотренных методик аппроксимации оценки точности прогнозов была выявлена наиболее эффективная методика. The most effective approximation technique was identified by comparing the considered techniques by criteria for accuracy estimation.